« June 2001 | Main | August 2001 »

July 2001 Archives

July 4, 2001

The only question that remains is "which of them do I fire?"

I went to Sandia with Andy this week.

It was a neat trip; I've been to Albequrque (sp?) before, but not to Sandia itself. I didn't realize that it was physically located on an Air Force Base (Kirkland AFB? Don't remember the name offhand).

Next time, however, I'll be flying in and out of Louisville, not Indianapolis. This time, I picked up Lummy in Bloomies and then drove to Indianapolis to fly out. We came home on Monday night, and after driving Lummy back to Bloomies, I didn't get back to Louisville until about 3:30am on Tuesday morning. Never again.

The moon and Mars were really bright the whole way home, though.

I also found out that you will get stuck behind someone slow on IN-46, regardless of the time of day.

The trip itself was cool. We went "behind the fence" at Sandia, into Classified County. We had to be escorted and within sight of Rich, our contact, the entire time. When we went into his office building, we had to sign in, and the secretaries put big magnetic "Caution -- Uncleared personnel in the area!" stickers on all the doors.

We were down there to see Brian and kick off our LAM/MPI collaboration with those folks. They actually can only generally tell us what they are using LAM for -- "simulating the nuclear stockpile". Any specifics beyond that are apparently on a I-could-tell-you, but-then-I'd-have-to-kill-you basis. Which is ok; I think I'm comfortable not knowing. :-)

Brian and I both gave talks; I gave a general "LAM is great" talk, and Brian gave a talk about specifically what we are doing with Sandia (we were up until 1-2am working on his slides, and then Brian came back to our hotel at 6:30am to practice his talk). Both talks went well. We met with Ron and the CPLANT folks as well; we'll probably be at least coordinating with those folks during this work, which is good.

In general, the trip was a success, and we have a better understanding of what they are trying to do, and what they would like us to do with LAM/MPI. This should be a very interesting project.

I notice the news reports that the European Union has blocked the GE / Honeywell merger. This was pretty much expected, I think --
there has been rumblings about this before. But it's still amazing. I am not familiar with the details, but I can't imagine that the US government is going to be happy about this (that's how much clout GE has). Even President Bush has apparently made some comments about how he is not pleased about this.


Tracy and I went to Gina and Dan's 4th of July party last night. There were lots of GE folks there. We stayed until after midnight sometime; it was quite fun.

My conversation with Rich about shared library modules for LAM continues. He brings up good points about the efficiency of shared libraries and how it's not the fault of the concept of shared libraries that, etc. And he's right. There's lots of very good technical reasons that we should use shared libraries for the module design in LAM.

But we won't.

At least, not right now. Right now, the state of technology for shared libraries (IMHO) is too non-uniform. Probably the number one reason why is that it's a nightmare to create shared libraries on different platforms. Even GNU libtool isn't a complete solution (it doesn't work on all platforms). Hence, LAM/MPI is not in a position to require shared libraries. Sure, it can be an option, but not a requirement.
This reason is closely followed by the fact that it would be a somewhat large delta to change to explicitly use dynamic libraries (properly) with dlopen() and friends. Is this rocket science? No. It's not even particularly hard. But it's a nonzero change, and, at the moment, unnecessary. So good engineering dictates that we don't do it now. Get it working (with static linking), and then possibly move to shared library modules. With good modular design, the change is the same now as it would be to do it in the future, so not implementing it now reduces the number of variables that we have to debug.

One point that came out that I assumed was common knowledge was that LAM can be compiled as shared libraries (using GNU libtool). Hence, all this module stuff can end up in liblam.so (vs. liblam.a). This was actually one of Rich's points -- that we shouldn't close the door on shared libraries completely, because of the various performance benefits, particularly for large SMP boxen.

My only point is that I don't want each of these modules to be their own shared library. Whether or not liblam is a shared or static library, I don't care -- that's a user choice -- just as long as it's not a requirement. More to the point, if the compiler/linker/libtool can give me shared libraries for free, great. But I don't want to explicitly program for shared libraries (dlopen() and friends) for the reasons that I stated above and in prior journal entries. At least not yet.

And, for the record -- I was right: Darrell did feel the urge to say the same things that Rich did. :-)

On another LAM note, someone found a minor bug in LAM 6.5.2 such that compiling programs with MPI I/O won't work. Arrghh... This may trigger the release of LAM 6.5.3. But we're currently in the process of figuring out what the license for LAM/MPI will be down at IU. I'm trying to push BSD, but other forces are at work (including IU's lawyers). We'll see how this shakes out...

July 8, 2001

Kenau Reeves can't act

Did you ever notice how green looks absolutely nothing like red?

I got my 9 "free" CDs from BMG (had to pay shipping and handling). They suck. It was damn hard to find 9 CDs in their selection that I wanted. And some of them I didn't really want. Ugh. And I think I ended up with at least one Sara McLachlan CD that Tracy already has. Double ugh.

The script that we've been using for CVS diff mails doesn't handle binaries nicely. I discovered that when I checked in a powerpoint presentation the other day -- it sent the whole binary file in the e-mail.

I hacked it up a bit so that it doesn't do this anymore. Apparently, CVS won't tell you if a file is binary or not. So I had to add a list of filename extensions such that the script will check the incoming files to see if they match. If they do, no diff.

I sent the script back to the Vorbis folks (that's where we got it from in the first place), and they put it in use immediately.

Tracy and I went to a 4th of July picnic with the people on our street. Nice folks. Met a few of the kids, too. One guy (retired) worked in a tobacco factory for 35 years as a machinist. He had some interesting stories.

A fairly serious (but pretty small) bug was found in LAM's mpicc this past week such that it was necessary to release LAM 6.5.3. I rolled in a few more minor bug fixes as well -- most of them had to do with ROMIO (the bane of my existence).

I had a long and exhausting conversation with Trond from RedHat about building the RPMs. As a result, I slightly improved the spec file for LAM, and we finally violently agreed that having the LAM RPM built on a RH 6.2 machine means that the man pages and doc files will be installed in the wrong place when the RPM is installed on RH 7.x machines. If the RPM is built (or the SRPM is rebuilt) on a 7.0 machine, since we didn't hard-code any file locations, the man pages and doc files will magically end up in the Right Place.

I think that I finally have the gm (myrinet) RPI for LAM working. I found many, many bugs, and a bunch of things that just weren't implemented yet. It passes the entire test suite on the hydra, Babel, and Chiba City (finally). I didn't realize that the test suite was so thorough. Let's hope that it is really thorough, and it found all the bugs.

I had to patch up a lot of the "make dist" procedure to build the new tarball, especially the lamtests suite (because lamtests now uses its own automake build procedure).

I'll wait for a few more confirmations (mailed it to Brian, Joe, and Feldy); hopefully we'll release the beta early this week.

A /. article caught my eye this evening -- some guy was mad because he got cut off from Telocity DSL when Northpoint went out a few months ago, and Telocity just got around to asking for their gateway back (they send you a pre-paid airbill box, by the way). Admittedly, lousy accounting on Telocity's party -- Northpoint shut down quite a while ago. However, it might still be in the 6-8 week range, which is fairly normal for these kinds of things.

The /. article played up the fact that Telocity charges $500 if you don't return the modem.


It's right in the contract. And it only makes sense -- if they give you a modem for free, it certainly makes sense that they want it back when you discontinue service with them. I know that I was either told that or I read it in the terms of service when I signed up with Telocity (both times). So I have no sympathy for this guy if he either doesn't have the modem any more or hacked it up to play with it.

A few people said as much. A few others disagreed. Idiots. How can you disagree with something that you agreed to, even implicitly?

So I posted pretty much what I said here, plus a link to the Telocity TOS web page where it specifically states that you have to return the modem or pay $500. I actually got moderated up to a 4 (insightful). Although I've really only posted on /. a handful of other times, this is the first time that I've been moderated up. Amazing. I posted fairly quickly, used bad grammar, and the posting wasn't entirely clear. But I guess I had the facts on my side.

Johnney Mnemonic is on TNT tonight. This movie sucks. I just noticed that Sandoval from Earth: Final Conflict is in this movie. His acting hasn't improved at all. So is Izzy from Starship Troopers.

The movie still sucks.

But hey -- The Planet of the Apes w/ Marky Mark is coming! Looks pretty cool; I always enjoyed those movies. It makes me wanna watch The Big Hit.

We had a big (but short) storm here tonight -- rain was coming down in sheets. It's a good thing that I watered the lawn today.

July 12, 2001

Jeff's Journal

Strange things are afoot at the Circle K.

In the slashdot post that I mentioned in a prior jjc post, Pete actually magically got moderator privileges that day (all the moons must have been in alignment), and so he moderated my post up to a 5 (the highest possible value). LOL!! Like I said, my post used bad grammar and was slightly unclear. I find it extremely amusing that it got moderated all the way up to a 4 (without help from friends), and then finally pushed up to a 5 (with a little help :-).

For those of you who are wondering, here's my post.

Good quote from Brian's journal:

"Friends don't let friends integrate ROMIO with LAM"

I picked up Janna from the airport the other day; they had just returned from a holiday of hiking through Switzerland. They had a great time and did much walking throughout various out-of-the-way little Swiss towns.

They bought me a swiss army knife with my initials engraved on it, and got some chocolate for Tracy.

We ran into Aimee in the Louisville airport; she was on her way to some business meeting the next day. It's a small world.

Eric fixed the latency problems on the Babel cluster. It seems that the NICs were continually going into auto-negotiation mode (to swap between 10 and 100Mbps) for some reason, which caused all kinds of retransmission delays and errors. But now the cluster seems to be working well (it's amazing that it has been that way for at least a year) -- the NFS delays seem to be much more normal, meaning that the latency is about what you would expect from NFS.

I drove back from Bloomies yesterday. Had a good 2 days there; I'll typically be spending 1-2 days a week there.

It's very easy to get from Bloomies to my house -- it's essentially 3 highways: IN46 to I65 to I64.

While I was driving home yesterday, I was driving along the forest-lined IN46 when suddenly 3 C-130's flew above the tree tops right in front of me (military transport planes). They were flying North. They looked to be at about the right altitude for parachute jumping. IN46 is nowhere near any military bases that I know of. Weird.

Later, I got on to I65 and headed towards Louisville. About 40 miles south, I saw 3 C-130s again (must have been the same ones, although they were much higher) flying east as they flew over me on I65. About 20 miles further south, I saw them again, this time on the West side of I65, but flying due south.

The Louisville airport was somewhat nearby, and they were heading in that general direction, but Ft. Knox is an additional 75-100 miles further south; they might have been heading there as well. Both airports can handle C-130s. Who knows.

The only obvious conclusion that I can draw from this experience is that the government is shadowing me, and/or bombarding me with electro-magneto kinetic rays in attempts to steal my brain. "This line is tapped, so I must be brief."

If I disappear and/or turn into a Microsoft neophyte, you all will know the reason why. Let the truth be known; trust no one.

I came across a great term yesterday: "war driving". It's a moniker off the old term "war dialing" from back in the 70s and 80s. War driving is taking a laptop with a wireless NIC and literally driving around and seeing what wireless networks you can tap into.

There's at least one windoze-based utility program for this: http://www.personaltelco.net/index.cgi/NetStumbler.

How secure is your network?

The SSI project in LAM is going quite well; I'm pretty excited about it. More and more ideas about what it can be used for keep occurring to me; SSI may solve a lot of issues and provide a really nice framework inside of LAM. If we do it Right, it may end up being the One True Way that we integrate LAM to all new kinds of systems (PBS/TM, Grid, Scyld, KLAT, etc.). Indeed, I'd really like to be able to use SSI to integrate new algorithms into LAM -- such as the tree-based lamboot. Very cool stuff.

I chatted with Kay at IU yesterday; she's a "pre-faculty" in the CS department (analogous to how I'm a pre-post-doc). Her advisor is Andrew C. (formerly of UIUC); she's part of the Fast Messages Group. We had an interesting chat about Myrinet, research, software engineering, Windoze development, etc. She's somehow connected to our IPCRES group, but indirectly. I don't understand the exact relationship (I don't think anyone does, yet :-). But she'll be at least somehow connected to our group.

It seems that one of the big things that Fast Messages did to make Myrinet message passing fast was to support essentially the same thing that writev() does (although they call it "scatter / gather", which confused me until I realized that she wasn't talking about collectives) -- take pointers to different chunks of memory and write them all out into a single transfer, rather than:

  • Force the user to copy everything into a single, contiguous buffer so that it can be sent in a single transfer, or

  • Force the user to use multiple transfers to send everything.

GM doesn't currently have such a vectored-write (or vectored-read) capability. So I pinged the folks at Myri about it and asked if they will ever support such a thing. Indeed, that would help us in LAM --
it would effectively eliminate the need for the tiny message protocol in the gm RPI.

July 14, 2001

I'm the original Nicks Superfan

My IP traffic to IU goes through Atlanta, then to New York, then to Bloomies.

My pilot fritzed out again the other day and reset itself for no reason. Last week, it fritzed out and lost all of its data (also apparently for no reason). This wasn't a tragedy, because I had done a backup in the recent past, but I did end up losing some data, which was annoying. This pilot is just getting kinda old; it was pretty hard to write on it (letter accuracy was pretty bad). Not that I'm an expert at palm writing, but when I wrote on Tracy's (new) pilot, my accuracy is much better. This leads me to believe that my (old) pilot is just getting tired. Indeed, the writing area is visibly worn.

So I went out an out a new m100. There's lots of other more advanced models -- indeed, m100 is the low-end offering -- but that's really all that I need. Although I had visit three stores before I finally found the m100 in stock (at CompUSA). Best Buy and Staples were both sold out.

I transferred over all my data and everything appears to be find with the new m100. My old pilot has officially been retired. It served me well.

So this brings the tally to five -- this is my fifth palm pilot. The record stands:

Pilot number Status
1 Death by rollerblading
2 Lost when left in the back seat of a rental car in Florida
3 Death by 3 foot drop onto asphalt
4 Retired at age 72 (palm pilot years)
5 Still in service

Tracy has broken a pilot as well (I think it was a short drop from her desk...?); she's on her second pilot. Now we both have identical m100s. Ugh...

I bought Depeche Mode's new "Exciter" CD while I was in Bloomies on Monday (stopped in Target to buy some Bennadryl, and ended up wandering over to the CD section). It's not bad, but the music is slower than their previous stuff.

Had a really long SSI call today. Dog, Brian, and I hammered out a whole bunch of stuff -- we've had separate discussions up until this point; this call helped have all three of us agree on a whole bunch of points. There's still a ways to go, but we agreed on things like:

  • There will be different kinds of modules (e.g., a comm module for ipv4 vs. ipv6, a boot module for the lamboot kinds of things, etc.).
  • Some kinds of modules will be use-only-one-of-all-available-modules (e.g,. comm), whereas some kinds of modules will be use-all-available-modules (e.g., the RPI).
  • The SSI glue will be general enough to not care where the modules are located (in liblam or libmpi).
  • The SSI glue will have three essential functions: init, finalize, and export_tables.
  • LAM's top-level configure will write out the tables that the init functions will use
  • Each module will have its own top-level interface that the rest of LAM/MPI will call. These top-level interfaces will do their own dispatching.
  • Modules will emulate C++ inheritance by specifying two global variables in each module: a struct full of a bunch of function pointers, and a pointer to a "base" struct full of a bunch of function pointers (which can be NULL).
  • Modules with NULL values in their struct of function pointers will have those NULLs replaced with real function pointers from their "base" class during init, so that during run time, we only have to do one pointer lookup, not [potentially] many.
  • The SSI finalize glue may not be called, but only if the process doesn't call kexit() (e.g., the lamd dies by signal)
  • The init function of each module may fail if that module determines that it should not run.
  • There will be command line interface that may be used to force the selection of a given module. It will likely have a three part nomenclature: a common option, the module kind name, and the specific module name.
  • Modules can call the SSI export_tables routine to get the final function pointer table for a given kind (probably with arg type (void*)) so that the SSI dispatch wrappers can be bypassed for performance reasons (this is important for the RPI).

I think those were the majority of things decided, in addition to the function breakdown of several of the [proposed] kinds of modules. One notable thing that we haven't decided yet is how to handle arguments of different types. The comm module has a notable problem: argument types and sizes are different between ipv4 and ipv6. We may get away with using handles to the "real" datatypes in many places, but Brian thinks we're going to have many problems in the routerd in the lamd, because it stores tables of IP addresses. Ugh. We'll see how that shakes out...

Connectivity to ND has sucked today. It keeps going away for 1-2 minutes at a time. Very frustrating...

I got /. moderator access this morning; I don't think that I've ever had that before. So I moderated a 5 articles and did my civic duty.

I went to see Tomb Raider with Tracy last night. It was fairly good. It wasn't what I would call a great movie, but it was worth seeing on the big screen and whatnot. Good effects, but I found some of the action a little hard to follow because they kept switching the camera at a high frequency. I'm sure that they did this on purpose, but I didn't like it much.

All in all, though, it was a good flick. Rimmer (Chris Barrie) from Red Dwarf was the butler. I give the movie 15 minutes.

Got some LAM patches from the KLAT folks to make LAM work on their systems. Their networks have the property that each node has multiple NICs, and the switches are wired up in a non-uniform way. i.e., nodes A and B may have entirely different ideas of what the IP address of node C is.

So LAM has to do something a little different than what it normally does: pass around hostnames rather than host addresses. Tim from the KLAT project hacked this up in LAM and sent me a patch. I'll probably be applying it soon.

Tim also noticed a small bug in LAM, such that when /tmp is NFS shared across multiple nodes, when the haltd in the lamd dies, it leaves the kill file open. This causes NFS to keep a cache file open (of the form .nfsNNNNNNNN) in the LAM directory, and therefore doesn't let tkill remove the entire directory. Blech. Tim's fix for this doesn't work when one compiles the lamd as separate pseudodaemons, so I passed it on to Brian to see what he can do with it.

I renewed squyres.com today for another 2 years (thanks for reminding me, palm pilot!) I renewed it a few months early, but hey... I don't wanna lose it. :-)


I have it on reliable sources that Nabisco is considering releasing a new animal for animal crackers since this year is the 100th anniversary of animal crackers. There are four animals under consideration; right now, Nabisco employees are being asked to vote for which they like the best. Supposedly, Nabisco will open the voting to the public later this year.

Here's the four animals under consideration:

  • Koala bear
  • Walrus
  • Cobra
  • Penguin

I think we all know what it should be. When Nabisco goes public with this, we'll have to post this to Slashdot and get all those Linux lovers to stuff the ballot box.

ND was supposed to switch to kerberos 5 this morning. By our syslogs (and the 6 bazillion cron e-mails that I got), AFS was out (on schedule) for several hours this morning, and came back a little after 6am. But checking several machines after they have been rebooted (they needed to be rebooted to get the new authentication scheme --
it's complicated), they didn't appear to be using kerberos 5.

What's the scoop?

Not surprisingly, the OIT has said nothing. Curt sent a "WTF?" kind of message to the AFS/Unix list, but AFAIK, there's been no response yet (it's about 2pm now). The OIT really sucks sometimes. Haven't they ever heard of communication? Haven't we been harping on their [lack of] communication skills for years now? Haven't they learned yet?

Apparently not. :-(

I just got paid by the military for my 2 weeks in Atlanta, but DFAS took out state taxes for IN even though I explicitly submitted a form saying KY. <sigh>

Some guy just posted a Solaris system administration question to the LAM list. Weird.

July 20, 2001

There is nothing more joyful than wallpapering behind a toilet

Ack. I rebooted before saving a lengthy journal entry. Doh.

The other day, I almost lost queeg. More specifically, I almost lost queeg's partition table. I was installing VMware to do some OSCAR testing. Wait... let me back up.

A long time ago, when I first got the box that queeg currently lives on, I originally had visions of dual booting it with Windoze. So I left a several GB partition on the disk for windoze. I even formatted it as FAT32, and mounted it under /mnt/windows in Linux. But then again, I never got around to installing/dual-booting Windoze.

Since I was running short on disk space (it's amazing how fast 20GB can get used up...), I decided to install my VMware virtual disks in the FAT32 partition. VMware complained that it either had lousy performance or couldn't lock the virtual disks or something in FAT32 partitions.

"No problem," I thought, "I'll just whack that partition and replace it with an ext2 (native linux) partition. I'm never gonna install Windoze anyway."

After a series of UTFS errors, I had gotten to the point where diskdrake (Mandrake's nice GUI partitioning tool) claimed that my partition table was corrupt and it couldn't read it. DOH!! After a morning full of fretting, backing up all my data, and carefully poking around with fdisk, I was finally able to restore my partition table and convert the FAT32 partition to ext2.

Moral of the story: it was the ReiserFS stuff that caused the majority of my woes (long story, I won't bother explaining here). While having a journaling filesystem is great, this ReiserFS stuff in Linux 2.2 can really bite you in the butt. I hope that it's better integrated in Linux 2.4.

On the up side, VMware is actually pretty nice. This is the first time that I've ever used it. I think I'll probably be buying a real copy of it (I only have a 30 day trial license right now) so that I can run windoze in that -- much easier than dual booting.

I really like the feature of their "undoable" disks. You can install an OS, mutz around with it, get it up to a known good state, checkmark it back to the persistent store, and then start testing. If your disk goes wonky, you can just say "throw out those changes --
let's reboot with the last known good state". More to the point, when you shut down the virtual machine, VMware asks if you want to commit all the changes that you've made to the virtual disk. If you say yes, all the stuff you did on disk will be visible the next time you boot that virtual machine. If you say no, the disk will be in the same state as it was when you booted.

Needless to say, this is extremely handy. It would be cooler if I could checkpoint the disk at any time (vs. only when I shut down the virtual machine), but there are some obvious synchronization issues involved there. Still, it would be handy.

I have 256MB of RAM in my machine, and that's enough to run 2 copies of VMware comfortably. Trying to run a 3rd at the same time causes major swappage.

Had a long chat with Darrell and Dian last night. It was good to talk to them again. Darrell is doing some very cool stuff at Yahoo. He has much more low-level kernel knowledge than I do; I really need to get into that stuff.

He also made a good suggestion about the versioning that we are planning on doing for SSI. It's amazing how we have lived thousands of miles from each other over the past decade, but yet our careers have managed to take many parallel paths. It's cosmic, dude.

D also suggested that I should have a way to get the titles from my journal entries in downloadable file that /bin/forture can use. It's a funny enough idea that I'll probably have to find the time to do that someday. :-)

The Code Red worm is running rampant on the net. I just have to laugh. It's a clever worm -- the authors made a few mistakes (like hard coding the IP address of whitehouse.gov -- duh), but the ideas behind it are both insightful and scary.

squyres.com has been hit with Code Red probes at least 20 times or so. kresge.com has been hit many, many more times than that. As of last night, www.cse.nd.edu had been hit over 58,000 (!) times just yesterday alone. (all the probes were ineffectual, because we all run Unix web servers, not MS IIS) D and I were wondering why my server had only been hit 20 times, yet others were seeing hundreds of thousands of hits. I wonder if the random IP address generator in the worm has a propensity for class A and B networks.

July 27, 2001

The great oepn source debate

I'm sitting here waiting for the MS vs. Open source panel debate to start (they're officially late starting, as of this point -- more than a few people have speculated that Microsoft hasn't showed up -- or "blue screened"). It's a big room and is slowly filling up (the center filled up immediately, of course). The Slashdot crew is sitting about 20 yards from me (Rob Malta, etc.), the Apache folks are right over there, Tim O'Reilly is strutting around the room. I'm sure that Miguel is around here somewhere, as well as various other open source luminaries. Quite a collection of people.

The amusing thing is that right before they opened up the doors to the room, RedHat passed out dozens of red plastic hats, and Sun passed out dozens of "OpenOffice.org" t-shirts.

So let's a lot of people wearing those hats, and at least a few wearing the openoffice t-shirts.

The panel should be interesting. Craig Mundie, the MS Veep who has been taking pot shots at the GPL and open source is the MS representative here today. He is countered by Michael Tiemann, CTO of RedHat. So Mundie is walking into an openly hostile audience. The first question that pops into my mind is "why would MS agree to this?" Do they really feel that they are right? Do they just want to show that they're not afraid?

I'm sure that news accounts of this debate will be all over .net... er... the net within a few hours. Might prove to be interesting.

I typed much through the speeches and panel. They're not guaranteed to be right (definitely abbreviated, and lots of typos), but I thought you might be interested. Some of the audience questions were pretty damn stupid, I have to say...

It's a day late, but you can deal.

26 July 2001
Shared Source vs. Open Source: Debate and Panel Discussion

Craig Mundie
Senier Vice President

Talk name: "Informed Choice"
- Our goal: creating an environment about informed choice
- My goal about speaking: not legistlate, create a dialogue, and inform others about long term ramifications of their choices

MS has no beef w/ open source. We think it's an integral part of ecosystem that has fueled tremendous success. But there are aspects of this movement -- free software and open source. And the press is certainly confused.

Mundie: open source isn't the issue.


Software ecosystem: industru -> customers -> govmt -> intellectual commons -> cycle


It's all about choice
- Develmnet choices: klang, community, source model, platform
- Distribution choices: open source, shareware, freeware, commercial and mixed
- Licensing choices: BSD, shared source licenses, traditional commerical licenses, GPL, public domains
- Business choices: services, packaged software, aggregate "distributions", appliances, hardware
- Each choice has policy implications


Software as a business
- MS choice: the commercial software model; built on a business model, licensing, investment in R&D, community and standards
- Software industry: an integral part of US economy, 148000 commercial companies, 2 million jobs, resulting in $28.2 billion in taxes paid, export revenues of $121B

--> He's driving that countries should be worried about this 'cause
free software doesn't provide jobs and income, and is therefore
unhealthy for


Learning from open source

- expanded community programs
- expanded source access: "shaored"source - range of licenses for different customers, partners and the intellectual commons, still provided under a commercial model


Summary: MS believes that the commercial model is good for the nationwide and global economy.

Michael Tiemann

"To be, rather than to seem" I claim: to build an arch of trust, it is better to be open, rather than to seem open. Same to be trustworthy, rather than to seem trustworthy.


He believes that the (?free/open?) source results in economic opprotunity. Cites fair/equitable competition.

Open source makes it much easier to be rather than to seem. Compares MS to alternatitive minimum tax -- which is neither alternative nor minimum.

Why would MS try this new high profile approach, when previous approach was working well?

Answer: Oct 31, 1998 -- the Haloween documents. There are a lot of smart people at MS. They see that OS model is a valid business model that can legitmately compete with MS.

GPL is the spine of OS. MS uses strong proprietary license. GPL is strong free license, like 1st ammendments. MS has benefitted (illegally -- his word) from application of its licenses. RH (and others) benefits from GPL protection. RH has always hit quarterly predictions, and went profitable a year early. Why is GPL bad?

Back to 1998. Revolution inside MS. Fueled by smart people in MS. Fueled by OS superiority. (cites purchase of Hotmail, tried to convert to windoze multiple times -- the light goes on and people realize "OS software is better"). "Do you think that the people who administrate those systems think 'Gosh, I wish that I could dump this BSD crud and replace it with windows?' I don't."

Shared source has nothing to do with building community outside of MS. It's not a licence, but a treaty, crafted by execs trying to buy time to quell the internal civil war.

MS has done much innovation. We are thankful for things like XML. etc. But "winner take all" attitude has to go.

When MS is ready to accept the GPL, and ready to accept fair competition (and many other comparisons), then we will welcome you to this party as a first class citizen.

The Panel

Clay Sherkie - acceletor group
Michael Tiemann
Dave Stutz (minme) - program manager for shared source
Mitchell Baken - chief lizard wrangler at mozilla (she wrote the
mozilla license)
Ron Johnson - attorney at ...., chair of internet law institute
Craig Mundie
Brian Belendorf - apache foundation

Tim O'Reilly: moderator

Craig Mundie: to respond. It's hard to know what to think by looking inside a company from the outside. There are many different views. We're not embarassed that people come forward and ask questions, whether we do the right thing or not. We have a single focus, though. The leadership is single minded in going forward. We have consistent leadership, though. Those who disagree can go do something else. Many of the ways that Michael characterized MS as "civial war" does not exist (at the management level) in MS -- nor at the rank-n-file level.

Tim: Brian, you've coem from the GPL side -- clearly, you've thought a lot about licenses and why. What are your thoughts?

Brian: Likes cycle slide -- research, gvmnt, industry, and users. But he thinks its bidirectional. It was important to us that lots of people use it, but to get people to contribute back -- to build community. Even though the obligation is not there (for the companies who use apache) to contribute back, people do. They understand. The creation of licenses and regimes that are bidirectional is what is missing from this debate. I think we'll see different level of input in shared source vs. open source. But then again, there are millions of MS developers out there, so we'll see.

Craig:I agree (bi directional). Giving code back is only one way of giving back. Giving taxes is another way (institutionalized).

Tim: How much does MS give in taxes?

Craig: First three days of WinCE shared source -- 10,000 people downloaded the source tree. We had a commercial kit for those who wanted it for a year now, we sold about 400 of those. We're happy about this. We give back financially, we give back in the stds world (XML, etc.). We will continue to seek ways to give back.

Tim: Dave: you're closer to the hacker level at MS, your thoughts?

Dave: No real war in MS. We're trying to learn good things from OS. There's a lot of people who have payed attention to OS. Sharing source is a who different beast vs. sharing binaries (from a supprot perspective). We're trying ot internalize that now. People do ask for more and more access to source code; it's become more central to people who development on a dialy basis. So now we're starting to develop these shared licenses -- it's a response to user requests. The standard that he's working on CLI/CIL is the same spec that Miguel is working on w/ Mono. Shows how short and simple the license is. I'd like to hear feedback, actually.

Tim: don't wait. Panelists, jump in.

Michael: I'd like to move from the nuance to substance. The efforts sound good. The logging companies are really nice as long as to let them cut down trees. Oil companies are very env friendly as long as they can drill for oil. But what about patents that prevent interoperability? So the substantial important difference is whether it is acceptable to -- where it is convenient -- to allow small parts of access?

Tim: But if the customers like it, who cares?

Michael: It's unfair; like civil liberties and crights, diff between those who make the rules and those who live by the rules.

Tim: ...missed... (general devil's advocate, siding w/ MS)

Mi: everyone needs full access to everything -- everyone can drink from the same water fountain.

Mitchell: the equilibirum that we have in the software industry today is flawed. The choice that is missing is the choice of leadership. Data and workflow is controlled by one entitty. That is not healthy for society, and for development of software, and not for the future. There's lots of smart people at MS -- but they get filtered thought the business vision of one company. We don't get to choose that from the lots of good ideas. OS should promote free software, leadership, etc. Characterizing OS as bad for policy is not healty; let's not let it succeed.

Craig: There is nothing on our part to characterize OS as bad.
<laughter> The ecosystem that we're working on is not just for cmputers. Other things as well. MS has very little sway with telecommunications, electronics, etc. We've had little success there.
<applause and laughter> But think about long term ramifications -- evolution of computing and ramifications.

Clay: issues of src code alone is less important than was 5 years ago. Meta issue is interoperability, not just OS. I'm more conccerned w/ open interfaces vs. the source code behind it. To Craig/Dave: In the Hailstorm documents, it says that there wou.ld be a wa for linux/solaris for them to participate. Can I use a hailstorm schema to have a palm pilot contact a linux server w/o a MS component in the middle?

Dave: Interop is key. There are a number of industries that have not seen the light and use all MS software. <laughter>
If the customers want it, we will make it possible. There is no question.

Audience: "Answer the question!"

Clay: This is not tru where in classes MS does not have a monopoly. I'm gonna re-ask the question and try to get a yes or no. --same question--


Dave: So...
<laughter^gt; Yes, but a caveat. As you know, in distrib systems, interesting things are done when parts you need are brought to the table. So pilot will want to authenticate and then talk to linux server.

Clay: Real question: is it a choice or a requirement? You're saying it's a choice.

Craig: It's historically been "the API". In a world we see coming, it's clear to us that you can't depend only within one machine. We don't believe convential stuff of RPC and whatnot. So in that world, protocols, schemas, and message packets are akin to API. MS has always published the API. The OS community has borrowed those APIs and made complete implementations. So when we publish the protocls and whatnot, anyone can do anything they want with them.

Tim: But we're worried about patents. Even though published, MS still has control. Even if not by knowledge/source/protocls, by law. Will there be patent protection?

Craig: We're a business. We license IP. If it turns out that this business says that we should license the patents, then we'll do that. But we are a business.

Tim: But Apple was a business when you copied their interfaces.

Brian : Still an issue of centralization. E.g., DNS. There's root servers. It's now privatized. The fact is right now, this is a critical point in the infrastructure. And we're concerned about it. Similarly, we're concerned that MS will control hailstorm, etc., etc. Worried that same type of centralization will occur in .net services. What draws people to OS is going away from centralization.

Craig: 2 things to think about. Right now, we're saying "This is what we're going to do". We've advertised what we're going to do. Downside: magic carpet/AOL. What is that? Hence, it isn't clear to me that we are granted automatic franciase in this area.

Tim: come back to point of health of overall ecosystem. Big concern about MS that you see yourselves as a small player in a big world. You think that you don't have power. But you do. In the ecosystem, finite resources. Makes it hard for new entrants to compete. OS says "we want in -- we want to have a chance". You guys have been so successful that its hard for others to succeed. So is proprietary vs. open "what's good for MS is good for the industry?"

Dave: Is it hard to enter to the market 'cause the expectations of users have been raised?

Tim: My exp is that it's easy to enter, but then MS comes in and takes over that part of the market.

Dave: I've seen repeated failures inside MS. We are not automatically granted the franciase.

Mitchell: But MS has plenty of money and backing to fail. But MS has efficient system to take $$$ out of the system (giving away browser, making it part of the OS, etc.). So ability for MS to extract $$$ is dangerous, OS allows people to collabotare together and joint together to be a larger whole.... missed...

Dave: We recognize that we are in a possition in which we have a lot of resources, and people are sensitive. We have started to try to be very clear about what we want to do. We can to carve out safe places
-- a standard. We need to continue to develop ways to make businesses in free markets. To continue to exist, but to foster trust and inniation.

Tim: So is it fair analogy to MS is switching from hunter/gatherer to agrarian?

Dave: I'd like to think yes, but can't say for sure.

Craig: MS would be nothing if millions of people didn't write apps for Windoze. And OS by itself is nothing.
Need a symbotic platform between apps and OS. Otherwise, OS won't sustain itself. So in a way, we were already a farming econ. In a way, we needed them. So people have diversified in number and type of platforms -- lots of diversity -- there's no direct transference, I can personally speak for our lack of ability to other non-computer systems.

Tim: Is this why the GPL bothers you?

Craig: No. Because the GPL makes it's own closed community.

Tim: So oes MS.

Cr: If the GPL wants to explain how to stand on their shoulders

Br: The GPL tells me under which terms I can use software (as a business). The WinCE license does not tell me anything as a businessman -- it says "contact us" for business terms. It's only for non-commercial stuff. Vast majority of people write software for some commercial purpose.

Cr: Just call us. We'll figure it out.

Br: But that's different.

Dave: Not only is there the noncommerical license, there is a community-based shared source license.

Ron: perspecive of a lawyer. The legal business has only recently entered this arena. Most of the commericla businesses have made licenses tailored to their best commercial business. This is not just an MS problem. Trouble w/ GPL to those doing commericla business is that expressed about 4 ways, none of which are very specific. Some are consistent w/ derrivitave copyright law, some are not. There is very little guidence from the courts as to definition of "derrivate work". There is no useful test to know. So there's a huge uncertainty w/ the GPL.

Br: Stallman is working on version 3. I'm sure that he'd welcome MS.

Crai: We've posted 20 questions on the web.

Michael: MS shared source, It has stipulations about IP. If I look at MS software, I'm infected.
It's the same language in both licensews (MS and GPL).

Ron: It's context sensitive. There are different language and different contexts.

Tim: Wrap up for audience. There are similarities between both licenses. Boths strong IP licenses, and probably both have ambiguities. So lets not go down the legal hole, 'cause we don't really know. Let's take a questions.

FSF president Q: MS stated that GPL is unamerican cancer. Yet this ctry was founded on freedom. GPL is founded on freedom. This has inspired free software movement. We'd like to challenge Mundie/MS to a second debate w/ the authors of the GPL to debate the philsophy of the GPL.

Cr: I'm willing to discuss it.

FSF: Oct 10 conference in DC. Will you be there?

Cr: Call Rick Miller, we'll talk.

Tim: This would be an interesting debate.

Cr: Richard wouldn't join our conversation. Why? Never mind. Next question.

..somedude...Q: To Craig: Ecosystems and choices. Then subject ofp atents came up and then dropped. There's little doubt that many patents are silly. But MS holds a lot of them. Do you think it's right when an OS infringes on a "silly" patent that they are persecured?

Cr: Absolutely.

Q: Even if it's a stupid patent? IT takes money to challenge a patent.

Cr: Fine. Get your money. Patents are one form of protection. Been debated intellectually, legally, etc. Should we have patents? Our society has said yes. Some are stupid. But they have legal weight. At the end of the day, if you have a ptent, you enforce it, it has value. So MS and others who have patents will decide to enforce them.

Br: Are there any for .net?

Cr: I expect that there will be.

Dave: I think that patents will benefit OS as a structural thing. You want large corps involved, ...

Michael: We want our turn.

Dave: To keep clear relationships, we need clear language and protection and way of litegation. Patent is outside of source. You need to help us to let us know what to exist.

Craig: Look in academia. Lots of them have patents and made $$$ off them.

Q: Steve Balmer said Linux is the biggest danger. I've been member of MS community since DOS 3.x and MSDN when it first began. Been a member of OS since win95. I understand OS "open" and "free". We beleive in open dialgue and open markets (and free beer!). When MS recently started using "community". CIA is part of "intelligence community". What does MS mean by "community"? You said earlier that there was freedom in the MS community by those who disagree they can leave. What does MS mean by "community"?

Cr: MS has lots of communitys. Employees. Developers. Customers. My comment (BTW -- we've used "Community" for a long time) earlier was that we can address needs better this way.

Q: Software ecosystem. As Tim and Mitchel pointed out, in traditional market of cash cow, you have an overwhelming position. It's difficult to make an annual growth revenue of 10-20%. Exp growth has limits. Is it time for MS to declare 0 revenue growth.

Cr: I don't think so. We have shareholders. Our job is to provide return to shareholders.

Q: Do you also have responsibility to ecosystem?

Cr: Of course. We try to be good corporate citizen. We try to provide good infrastruct for this country. We try to step up and deal w/ corporate responsibilities.

Mark Brickell EFF Q: I'm just a sysadmin.
<cheers and laughing>
One of the issues that concerns me is that we're in a monolthic society. Oil, media, power, etc. Only thing that seems free anymore is software. In that battle, you community has taken one of us hostage -- Dimity. How does MS feel about enforcement of DMCA? Are you going to keep taking hostages for free expression and free speech?

Cr: I suggest you address your question to Adobe.

Q: Did MS lobby for the DMCA?

Cr: We talk to people to all the time. The DMCA is what it is. It's the law of the land. Go change it if you want.

Tim: Does MS like it?

Cr: There's some we like and some we don't, like you.

Tim: There's very little that I like about it.
It goes too far.

Q: Ecosystems. As a biologist, the width has to do with its health. Narrow ecosystems (monocultures), you have to do things to keep them viable. Pesticides, etc. We've got a forest of all the same type of tree -- one parasite can destroy it. What's the vision of MS and other firms -- how do you protect them from this one parasite?

Dave: It's imp to our business to have a healthy eco.

Michael: Buy RH Linux.

Q: I interest of brokening peace between parties. michael said, "MS shouldn't be winner take all". But GPL is a "winner take all" strategy. Stallman says "eliminate competition". But GPL has potential to destroy (weaken?) ecosystem by creating a monocoluture. Why not offer something to both camps rather than either extreme?

Tim: This is a loaded hot potato. I think that the univ licenses are the best balance between freedom and making $$$. At the same time, I respect and support the right of MS to put out and make $$$ under thier liencese, and I respect GPL. It's "what works for you". If their customers don't like it, there will be choices. We are entering time of more choice -- because of new technology and OS. We can make the future what we want it to be.

Tim: We're out of time. If you want to ask Craig more questions, go to the Oct free software event. Thanks.

July 28, 2001

Open Source Convention summary

These are my thoughts -- not heavily edited, nor intended for publication in a "real" outlet (this is a pretty standard blog disclaimer). Heck, there's probably still lots of typos and grammar mistakes -- I read it once after I wrote it and made a bunch of changes, but I'm sure it's not perfect, nor would I expect them to hold up to rigorous scrunity. But they're a summary of my thoughts, and perhaps you'll enjoy them, and/or have something to think about after reading them.

The conference was in San Diego, a city to which I've never been before. The airport is actually in the city -- it's weird to fly lower than some of the buildings while coming in for a landing. The meetings and whatnot were spread across the two buildings of the Sheraton right near the airport. Walking between the two buildings was a nice five minute walk along a marina. Great weather provided some nice scenery between sessions/meetings.

There's great wireless connectivity in just about all the sessions. All the keynotes and breakout meetings had good wireless coverage -- even walking between rooms worked nicely. It's been really handy to have internet connectivity during some of the talks. Indeed, I transcribed most of the open source debate w/ MS's Craig Mundie and immediately mailed it off to a few friends and internal mailing lists when it was done. I even managed to stay not too behind on my e-mail -- lots of LAM mail that I didn't even get to this week (including some from Brian; oops :-\ ), but I did manage to handle a bunch of other stuff.

I have to admit that I found this conference more interesting than I thought I would -- I've actually gotten something out of this conference other than "open source is great, we should all use it." If nothing else, it was refreshing to see the enthusiasm of all the young (and not so young!) coding punks out there, and talk with them about what they were doing. It was also good to see that there are some smart, Important People wondering about the future of open source, and how to keep it alive.

That being said, there was also a bunch of the predictable "Microsoft should die" kinds of things, as well as "all patents are bad; they should be abolished" and "all software should be free". While I personally have nothing against these kinds of zealots, and, indeed, I may not totally believe their positions (and sometimes not believe them at all), I do understand where they are coming from and am at least somewhat sympathetic to them.

However, (IMHO) such zealots tend to ignore certain realities in their ideological zest. Someone recently said, "Some things just are. The speed of light for example, while I personally think it is too slow, is unchangeable." Ok, so there are few things that are as totally unchangeable as the speed of light. An example: getting congress to abolish patents is effectively trying to change the speed of light; there's no money in the open source movement to do so --
more importantly, there's too much money in the opposition that supports patents such that congress would laugh at the very idea of abolishing patents. Heck, they wouldn't even laugh, they'd totally ignore the prospect. It's an impossible task.

I am not such a pessimist that I believe this for all aspects of things that I personally believe to be undesirable, but I do certainly believe that one needs to pick and choose their battles. Abolishing patents is not a battle that we should pick (nor do I really believe in that one -- I'm just continuing my above example).

Does this make me a cynic? I prefer the term "realist".

I do believe that there is much that we can do. I believe that the work that we do (writing free software) is important, and that we can accomplish some really Great Stuff and help people in the mean time. Especially as an academic, we can produce Quality free software through the open source paradigm and still get funded (as is all-important in academia) while still in the university spirit of open intellectual research. We do what we do because a) we believe in it, and b) we enjoy it. That isn't going to change for the foreseeable future.

I arrived here Tuesday night. I registered at the conference and went to the Sun Grid Engine BOF (birds of a feather session --
typically an informal meeting to discuss a given topic). It was pretty much an ad for Sun's product, so I left and went to the CVS BOF. I specifically went to that BOF because we use CVS every day, it's a great tool. That being said, it sucks. There are some well-known problems with CVS that can be really annoying. CVS is kind of a "best of the worst" that is sometimes a common label among free software. :-)

My purpose in going to the CVS BOF was that I wanted to see what the future of CVS was going to be. Indeed, Brian Behlendorf from Collab.net was there, as well as Derek ?Atkins? (the current man-in-charge of CVS, so to speak). My take from the BOF was that CVS is loosely supported by Collab.net and Derek, and there are a few new features that are planned / halfway written. Over time, these features will be added to the mainline CVS distribution. However, they really want others to help contribute to CVS; they're not spending huge amounts of money and/or time on developing CVS. A bunch of the Bad Things that we don't like about CVS won't really be fixed, both for the reason cited above (Collab.net not spending a lot of time/money on it) and because CVS is somewhat architecturally limited such that fixing them would entail a complete rewrite.

Loosely speaking, the future of CVS is SubVersion (I didn't previously realize that the SubVersion people were the same as the CVS people). SubVersion will basically do everything that CVS does and fix all the known Bad Things about CVS by starting over from scratch. Apparently, SubVersion will soon hit a milestone where they start using SubVersion for their own version control -- this should happen by the end of the year. At that point, SV will be usable, but probably not very feature-rich, and not have any GUIs, etc. They expect to have a full, feature-rich SV in about 2 years.

Should be interesting. I probably won't play around with SV now, but maybe in a few months, when it gets comfortably past the "usable" stage.

Then I met Rich for dinner. It was great; I hadn't seen Rich in probably at least a year. We went to an Irish pub, had a feast of food, and talked about All Things Geek. Then we went to a Hookah Bar where we ran into two of Rich's friends. Rich rented a water pipe with some apple-flavored tobacco. It was interesting. I can now say that I have smoked a water pipe -- check that off my "life experiences" list. :-)

The next morning, I went to the two keynotes. The first was from Fred Baker, entitled "Will the Next Internet Generation Still Depend on Open Source?" It was somewhat doomy and gloomy, full of caution and worry. While I did not agree with all of his points, he did make some very good ones (that others in the audience certainly did not agree with). A good example can be summarized as, "When will my mother use Linux?" This is very true. Sun's usability report on Gnome that was published a week or two ago showed that even though things like Linux are pointy and clicky, they are still heavily geared towards geeks and not the general public. This is a problem. It certainly can be fixed, but it needs to be recognized first -- the open source movement has to "grow up" and recognize that things are what they are, and to get widespread adoption, we have to cater to the public, not the other way around. Very true. Another example that he cited is that customers want software that is stable. Features are great, but stability is necessary if you want to use software with a business. This is something that I think open source programmers are starting to realize ("stable" and "unstable" trees are becoming more common).

The next keynote speaker was much more fun -- he was W. Philip Moore from Morgan, Stan, Dean, Witter, and gave a speech entitled "An Open Source Success Story on Wall Street". He is a programmer for MSDW, and uses open source products all the time. His main point was that OS is great, and Big Companies like MSDW love OS. They even fund OS projects. An example (that I'm forgetting some of the details on) is that they needed some extra features in MySQL for their particular. So they paid MySQL a few hundred thousand dollars to do it. MySQL and the MySQL community wins, and MSDW wins (a few hundred thousand dollars is chump change to MSDW). Yay for everyone! :-)

Another of his points was that vendor support can suck. Just because you're paying someone support contracts doesn't mean that the support that you'll get is any good. Managers think that paying for support gives them a warm fuzzy fallback when things go wrong, but the reality of the situation is that this is not always so. The OS communities (in his experience) tend to give better support, and fix bugs faster than vendors. He was extremely happy with Perl and the Perl community over the years. Indeed, MSDW now uses large numbers of Perl scripts to run their enterprise.

So these were refreshing words to hear. And it brought to light a previously unknown open source champion -- the major corporations who don't necessary write open source software, but use open source software. Perhaps these companies may be able to be persuaded to take up the banner, so to speak. If open source helps their bottom line, what would happen if it all went away? (this is a much longer conversation, but it's an interesting proposition)

After that, I went to the O'Reilly Summit on Open Source Strategies (a track within the overall conference). Tim O'Reilly himself spoke, as well as some other well-known folks. It was fairly predictable stuff, though. The big question was how to make money off open source and/or free software. Selling support has frequently shown to be not enough (although not in all cases). Companies like RedHat are frequently held up as if to say, "See? Open Source and free software can make money!". When actually such companies are (at least today) more the exception than the norm.

Making money off open source / free software is a problematic issue. Starting from scratch with a new product that is both open source and free is a difficult position to make money. The ones who are making money, for the most part, were either established companies before they went open source (IBM, Sun, HP, etc.), or were deeply established products before they went corporate (Berkeley DB, Sendmail, etc.).

I don't know the answer to these kinds of questions; no one does yet.

I met some guys from the University of Arkansas at lunch; a sysadmin/instructor and two of his students. They were fun to talk to. After lunch, we all went to a talk from Dave from Microsoft about some of the infrastructure of .net and C#. Specifically, it was about CIL and CLI -- the meta language and run time environment for that meta language (I forget which is which) that will comprise the backbone of some of the .net stuff. It was actually fairly interesting; I think Jeremy/Todd/Ron/Jeremiah would have enjoyed the talk because it was about language design and compilers and the like. I got the impression that Dave was a manager, but still very much an engineer and coder, so he spoke the same language as most of the audience.

As with all things MS, it sounded great. It's totally vaporware at the moment (and potentially quite monopolistic), so I reserve judgment. But the technical side of it sounded pretty cool. Whether MS actually delivers or not is a different question. And more importantly, what is left out of what MS deliver? What were they not telling us? What's the catch? Historically, I have come to not trust Microsoft. It sounds great, but I'm not a believer. We'll see how it plays out.

After that was a talk from Miguel from Gnome about their implementation of the CIL/CLI/.net stuff -- Mono. He has a heavy accent (he's from South America... Brazil, IIRC?), and talks very fast. He's a funny guy, though, and pretty smart. His take on this stuff was that he thought it was very exciting, cool stuff. Their reason for doing it in Gnome is not to be compatible with MS (although that's a nice, desirable side effect) -- they're doing it because it actually provides them with good infrastructure for advancing Gnome. It gives cross platform, cross language connectivity, and a reliable and modular approach to the software engineering of a complex system.

Both talks were fairly interesting. It'll be interesting to see what Mono is not allowed to do -- I find it hard to believe that MS will allow them to be first class citizens in .net. We'll see how it plays out.

I went to the beginning of "The Challenge of Privacy and Security" back in the Summit part of the conference. I think the best quote was from a Ph.D. who is the head of a privacy watchdog, that went something along the lines of (ok, I'm totally paraphrasing. Cope), "However bad you thought it was in terms of privacy on the internet, it's actually 100 times worse."

She told horror stories of how companies are gathering and cross referencing enormous amounts of statistical data on web surfers, 99%
of it without the knowledge or consent of the user. Such information gathering is typically not done out of malice or a desire to control users, but to help their marketing -- companies don't think that they are doing anything wrong.

Another assertion that I have privately believed for a while is that initiatives like TrustE have failed. There have been well-documented cases of companies that were certified by TrustE (or one of the other companies like TrustE) who then reversed their position and started (for example) disseminating private information that they had collected on their users. The long and the short of it is that paying for a seal of approval is just that -- it doesn't necessarily mean that you conform to the guidelines of what is implied by that seal of approval.

Then I left and went to a talk from LLNL entitled "Steering Massively Parallel Simulations under Python". I went specifically because these are the guys who who pyMPI -- Python bindings for MPI. I talked to Pat, the main author for a little while afterwards. Their main purpose for doing it was to do rapid prototyping -- they could hack something up in python and play with various algorithms before handing it off to a computer scientist to code up in C for real production runs, etc. He said that coding up in python was much quicker for them than coding up in C, and so the net time saved was actually very large. Interesting perspective.

We also chatted about using pyMPI with LAM (it does compile and work with LAM, of course). He said that pyMPI was initially developed using MPICH, but then as more people wanted to use it, they added LAM to the "supported" list as well, and it was a good portability lesson for them. MPICH makes different assumptions than we do, so expanding their code to make it handle LAM as well was good for them (his words [paraphrased], not mine).

After that, I went to dinner in the big open-air tent between the two hotel buildings. I picked a random table and sat down with some people that were already sitting there. Most were business types of one flavor or another (programming consultants, independent small software companies, etc.), which provided some interesting conversation. The more interesting guy was Michael, who is finishing would could be best described as a "walkabout", in the true Australian sense of the word. He got married and three days later he and his wife started a would tour that went wherever their feet led them.

He was a programmer before he started his walkabout, and he and his wife are now winding down the grand tour, and thinking about stability somewhere with a real job, etc. So he decided to attend this conference to catch up on the current State of Things. Great quote from him about why it was time to end his walkabout, "We were in Nice a few weeks ago, and were thinking, '<Yawn> Another fucking beautiful cathedral. <yawn>' Yeah, that's a good sign that we're done."

He was interesting to talk to.

I went to the Slash BOF after that. More specifically, I was wandering by all the BOF rooms when I heard some voices that I recognized. It took a moment or two, but then I placed the voices: I had heard them in "Geeks from Space" installments on Slashdot. So I wandered in and sat down to listen to what they had to say.

Rob Malta and several of the other Slash and Slashdot crew were in there. They're an arrogant-yet-funny group, and have lots of inside jokes with each other. They reminded me of any good programming crew. They talked about some of the upcoming features in Slash (and therefore, someday in Slashdot). They also had a bunch to say about the optimizations that they code for (have to be able to handle an absurd number of page views every day), and the extensive infrastructure that they have behind the scenes to handle all the user traffic and thwart evil/asshole users. Pretty cool stuff, actually.

Andy came in really late that night; the plane that he was supposed to catch left 10 minutes early for no apparent reason. He and about 10 other passengers were left behind. So he had to catch a later flight. Weird.

We went to the Great Open Source Debate keynote the next morning
-- "Shared Source vs. Open Source: Debate and Panel Discussion" with Craig Mundie from Microsoft (MS), and Michael Tiemann, CTO of RedHat (RH).

I've already posted my version of the transcription of the speeches and debate. I should have done it immediately instead of waiting a day (indeed, I only sent it to a few friends and internal mailing lists), and I could have gotten slashdotted. ;-)

Andy and I chatted about the debate afterwards. He's my take, with some flavor/discussion from Andy as well.

Microsoft won. Craig Mundie calmly discussed his position, and was in an easily-defendable position. Michael Tiemann and [most of] the various people who asked questions from the audience came across as whiners, saying "It's not fair!"

Come on, folks -- Microsoft is in the business of making money. Why on earth would they change their model when their current model is working very well, and doesn't show many signs of abating? Their bottom line is to a) make money, b) increase shareholder value. That's it. Call it malicious, call it evil, call it whatever you want
-- it's business. Yes, they happen to be the biggest software company in the world and have tremendous influence, but (whether we like to admit it or not), they earned that position. Yes, with flawed and crappy software, but people bought it anyway. Regardless, that's not what we're debating here.

Remember: [potential] technical superiority of open source is neither a necessary nor sufficient condition to debunk MS.


Don't get me wrong; I think that many open source products are vastly superior to MS products in many, many ways. But just because something is open source does not make it superior to proprietary/MS products. There's a lot of open source shit out there. Troll around on freshmeat, where every teenager who has every written a shell script has "published" their "software package". 98% of them are total crap and only work on the machine that the author wrote them on, or only work on Linux (which, to me, is useless).

There's a helluva lotta software engineering and design that has to go into a successful, Quality product. Just because you're open source and/or GPL doesn't mean that you are Better. So when I talk about technical superiority of open source products, I'm talking about the big Quality products, like Apache, MySQL, Postgres, PHP, etc., etc. Not every tiny little open source project out there.


MS is successful for many reasons. But the fact is that they are very successful. They have an enormous percentage of market cap. Why on earth would they want to give up even 1% of that to open source? Of course they're going to fight. Of course they're going to doublespeak and claim that they are better (technical superiority). Of course they're going to say that GPL is anti-business. Of course they're not going to play nicely with others; that would be giving away market cap. Of course we're going to feel like the underdogs (which, in many ways, we are), and feel that this is not far. This is not under debate.

To defeat your enemy, you must first understand your enemy. So understanding MS's position is important. Call it greed, call it business, call it whatever you want -- they're making money and they're good at it. Understand that, and then calmly, rationally proceed. Bringing up religious arguments (software should be free, you guys suck, etc.) is not helpful, because like it or not, MS has the law of the land behind them right now. They are are in perfectly defendable position of saying "the law agrees with us" (I'm not talking about their potentially illegal monopolistic practices here --
I'm talking about their attitudes towards open source/free software and the fact that they don't feel that they need to be compatible with anyone else, and their "embrace and extend" attitude). You may not like that, but it's a fact (similar to RIAA issues with napster and whatnot). Like I said above, it is what it is. Go read The Prince.

The next step is to figure out what we're going to do about it. How do we a) make money as well, b) eventually displace Microsoft and/or force them to make higher quality products and place nicely with others? Reasoning with them won't work, because the almighty dollar is always a more persuasive argument -- trying to reason w/ MS saying, "hey, give us some market cap and then we'll think you're good guys" is simply not a compelling argument.

Indeed, some audience member asked Mundie to sit down with the GPL authors at some upcoming FSF/open source convention on Oct. 10th of this year -- the audience member said that if MS's beef was with the GPL, they should be debating with the GPL authors, not open source luminaries (they are different things, for those of you who don't know). Mundie replied, "Richard wouldn't join our discussion" (referring to the fact that MS tried to initiate license discussions previously). Hmm. That's quite damning, actually. I can imagine why they didn't (because it wouldn't have been constructive, odds stacked up on MS's side, etc.), but it comes across as "the GPL folks will only join the conversation when it suits them". Indeed, Mundie came to the open source convention where the odds where stacked up against him, didn't he?

Some have criticized Mundie for the following exchange during the debate:

Audience member: ... Do you think it's right when an OS infringes on a "silly" patent that they are persecuted?

Mundie: Absolutely.

Audience member: Even if it's a stupid patent? It takes money to fight patents.

Mundie: Fine. Get your money.

Mundie's position here is perfectly defendable.

  1. Why should MS be punished when they have anted up the money to defend patents and whatnot?

  2. The abided by due process; it's too bad that others don't have the money or resources to do the same, but it's not their problem.

  3. They paid their dues, built up their company from scratch, and now have the resources and ability to do such things.

  4. It's the law.

Is it annoying and contentious? Yes. Is it fair? No. Is Mundie right? Yes, he is. Democracy isn't fair. Neither is business. Capitalism does not equal freedom.

Our country is built on change. That change has always had rules associated with it about how to acquire that change. Right now, those rules involve a lot of money (lobbying in D.C., etc.). I don't necessarily like that fact, but try living in an oppressive third world country where people daily have to fight for food and then tell me that the US sucks.

Michael Tiemann came across as a whiner. "It's not fair", "We just want our turn", "Buy RH software". Shut up. That's not the point. Indeed, what would happen if RH was in MS's place? Would RH still be so altruistic? Who knows. The fact of the matter is that I wouldn't want RH in MS's spot -- I don't want any one company to be the top (this is one of my big beefs with MS, incidentally). I contend that any firm that enjoys a monopoly like MS currently does would act exactly the same way that MS is currently acting. There's a reason that unscrupulous bastards are king: Money talks, bullshit idealism walks. Specifically, I believe that RH would be just as bad as MS if they were King right now.

So I think Tiemann's approach was entirely wrong. He should have addressed the monopolistic practices of MS. He should have cited the illegal and underhanded activities (he sort of did -- his "be and not seem" remarks were right on the money, actually). Cite the concrete legal issues about what MS is doing wrong. Not whine about "it's not fair", and try to whip up religious fervor by drawing comparisons to Rosa Parks and free speech.

Some of the panel's most insightful comments came from Brian Behlendorf and Mitchell Baker. They came across as calm, insightful, and rational arguments. Much of the other stuff was fluff and religious fervor.

Perhaps the most constructive comments that I have heard so far in reaction to this debate were along the line of:

  • Just write excellent code. Keep writing it. Write the best damn code that you can. Even if technical superiority isn't a sufficient condition to win, it can certainly help.

  • If you don't know anything about the political process, let the appropriate people handle that. The politics of this beast are very complicated; let those who know how to play the game make the public movements. Stuff like "MS sucks!" doesn't help, and makes the rest of us look stupid.

There. That's my $0.02. Comments are welcome.

After the debate, Andy and I went down to registration to straighten some details out. Lo and behold we ran into Johnney (old timer in the LSC). Small world! We exchanged cell phone numbers and then ran off to the next session.

Andy and I watched a neat presentation from the Collab.net folks, and then a presentation from Sun's openoffice.org guy. Both were pretty good. The Sun guy had three important messages that he kept hammering throughout the speech (I hope I got these right!)

  • Open source is a lot of work
  • Open source is not free
  • Armadillos don't make good pets (arrgh... I have no idea what his third point was)

These are good and important messages. Sun spent millions of dollars converting the StarOffice tree to be open source (translating it to english, cleaning up the code, going through all the legal hassle of ensuring that they owned all of it, etc., etc.) which many people don't seem to realize. "Just publish the CVS tree!" isn't an easy thing to do.

Nothing ground breaking in these sessions, but they were good talks. We talked with the Collab.net guy after the sessions, and we might end up using some of their tools for our own work back at school.

Andy and I wandered to the open air tent for lunch. We grabbed some box lunches and were looking for a table to sit down at. We walked by a table where Tim O'Reilly was sitting, and I made the offhand remark, "Wanna have lunch with Tim O'Reilly?" We walked about 10 steps further and then Andy said, "Sure, why not. I didn't get to ask my question at the debate this morning."

So we sat down next to Tim. He was in the middle of a conversation with some other people, so we just sat and listened for a little while. Then I caught his eye and we introduced ourselves. We had a short chat about the role of universities in open source and the whole movement thing (Tim is a big believer in the participation of universities, incidentally). He had to run off to handle other things, but now I can say, "I had lunch with Tim O'Reilly".

We chatted with the other folks at the table too. It turns out that two of them were academics as well (Carnagie Mellon), and they had very similar views about software as we do, which is very rare in the academic community (i.e., software should "just work" and suck less, that it should work on all platforms not just linux, etc., etc.). I didn't catch their names (more specifically, I don't remember them), but I think they were the primary people behind the Festival open source speech synthesis software. There was another guy from Cisco at the table who was having animated discussions with them as well, so I think he was either interested in, or involved with the speech synthesis/telephony over IP kinds of things.

Michael, the interesting guy whom I met at dinner the previous evening, joined us at the table as well.

Andy and I eventually wandered outside to the grass by the marina to discuss "stuff", including the morning's debate and whatnot.

We next when to an OSX talk. I'm not a big Mac fan, so I didn't pay much attention, and instead took advantage of the internet connectivity to catch up on some important pending e-mails, etc.

Andy and I tried to go to some extreme programming sessions after that, but the sessions were already jammed and overflowing with people, so we went to the exhibit floor instead. We ran into Johnney again and generally wandered around the floor. I only got 3 t-shirts. There was another free t-shirt from O'Reilly, but you had to buy the "Learning Perl" book, which I didn't really want to do (and therefore the t-shirt wasn't really free, was it? Hmm...).

Nothing too earth-shattering on the floor, but I did chat with the president of SAGE (System Administrator's Guild), who's a sysadmin from the University of Wisconsin. He knows Curt (of course), and so we had a nice talk about Condor, LAM, and general sysadmin stuff (I amused him with one or two Army sysadmin stories from Atlanta). I might well sign up for SAGE; who knows how it might benefit me in the future (it's a tax break, too!).

He said that one of the things that they are working on is a standardized test for system administrators. I think that would be great. Indeed, when I first got to my army post in Atlanta, the whole network was a mess. And this was from a full-time sysadmin who they paid a good amount of money to. Here's one example: when my predecessor applied patches under Solaris, if he ran out of space in /var, he would NFS mount some other disk to finish applying the patches (if you don't know why that isn't a good idea, don't become a unix sysadmin).

We were a little late getting to a session back in the Summit track from Eric Schmitt (sp?), president (I think, or CTO, or CFO, or CIO, or...) of Google. He had some pretty interesting takes on open source and its ramifications. Well reasoned, well thought out, etc. A good speaker, and an intelligent man.

Fun facts about Google that he shared:

  • Google has 4 data centers. Each center has a huge number of computers (he said the numbers, I don't remember -- let's say 1,000 each). Each data center is hosted by a professional company, such as Exodus.

  • They stack their computers 40 or 50 in a rack, and have to put the racks as close together as possible "for speed of light reasons".

  • They use Pentium III/750 chips rather than Pentium IV chips because when they tried PIV chips in 40-50 computer racks, they would melt the ceiling tiles. And this is in a professional hosting site, where they have enormous air conditioners, etc. They had to bring in a cooling engineer to redesign the air flow, etc. So until the chips are cooler, they're limited to PIII/750.

I don't remember too many of his specific points (should have taken some notes...), but I do remember agreeing with most of them. We went up to chat with him afterwards. Andy asked him about the role of universities, etc., in open source. At some point I interjected saying something about how most academics only develop proof-of-concept quality code. He agreed, saying (paraphrased), "Most academics code in Java, C, or Perl for Linux."

"Oh no," I said, "We code for POSIX -- we think that our software should work under all flavors of unix."

He literally did a double take. "Really? That's very unusual in academia..."

So that's pretty cool. I made the president of Google do a double take. :-)

We were walking out of that session, on the way to meet Johnney for dinner when we heard someone say "cluster". We turned and it was a guy from Compaq. We butted into the conversation and learned about an interesting project at Compaq to make single system image systems for linux (also see here) -- but different than Scyld. Their idea is to take a cluster of machines and show the entire aggregate as one machine. i.e., all the processes, all the disks, all the RAM, etc. This is different than Scyld, who believe that the non-master nodes are just shells for computation. For example, if a process migrates around in one of these systems, the entire process migrates -- it doesn't leave behind proxies for system calls (sockets are still problematic, though). Indeed, we might have less trouble porting LAM/MPI to this kind of system than to Scyld systems (don't have the filesystem issues, for example). Could be interesting.

We joined Johnney, his cousin, and a friend of theirs who lives in San Diego for dinner at the Ruth Chris Steakhouse. Yum. Good food, good conversation. It was great to see Johnney again.

Andy and I went to an extreme programming BOF that night which was somewhat interesting. Much more interesting was a BOF entitled something like "When Politics and Open Source Mix". This was an extremely enlightening BOF.

It was hosted by some guy (I'm so horrible with names -- I wish that I could remember his) who used to be a white house staffer, then worked as a programmer for a defense contractor (Nichols Research -- a "beltway bandit"), and now does independent consulting. There were only a few random people at this BOF, which was a shame. It seems like there needs to be a lot of people involved.

His main point was that he has seen the political process from the inside. In general, it goes something like this:

  • Congressman Smith gets a complaint about X.

  • Congressman Smith knows nothing about X, so he turns to his staffer and says "Find out about X. Find out how we can a) resolve it, b) make me look good, and c) make our contributors happy" (not necessarily in that order)

  • So the staffer goes off and first calls all their contributors and says "How do you feel about X?"

  • The contributors weight in on how they feel about X.

  • If the staff is feeling adventurous, the staffer will try to find out more about X outside of just the contributors' opinions (but not always). This may involve, say, Google searches, some more phone calls, etc.

  • The staffer will take all these opinions, and reconcile them with the three goals set out by Smith (resolve the issue, make me look good, and make our contributors happy).

  • The staffer will present their report to the Congressman Smith, who will then act accordingly.

This is a big part of what happens -- making the contributors happy. For example, have you noticed how the Washington Senators and Congressmen are against the Microsoft lawsuit? Why? Because MS contributes lots and lots of money to them. It's all legal --
there's nothing nefarious happening here. We're not talking direct bribes (try not to be cynical), we're talking support for campaigns, creating jobs for constituents, etc. But the total dollar figures are staggering.
Money talks.

MS publicly admitted that they goofed w.r.t. the initial lawsuit back in 97 or 98 -- they had completely ignored D.C. until that point, thinking that it was irrelevant to what they were doing. They admitted their mistake, and have since rectified it. MS now spends enormous amounts of money every year lobbying in D.C. and elsewhere. Have you noticed how New Mexico just settled the lawsuit with MS? There were many reasons (new DA who inherited the case, etc.), but at least some of it was due to the fact that MS "made it easy" for them to settle by offering incentives and attractive terms to settle.

Our BOF leader's point: most congressmen and senators don't have an opinion about open source or the GPL right now. But soon enough, they will be forced to have one. Some court case will occur, or MS will address it with their own lobbying, or one of a hundred other things will happen that will force the issue. The question is: if MS is doing all the talking to the lawmakers, who is talking for open source?

It's an easy enough word association for the senator or the staffer -- "Hmm... open source software. 'Software' -- I guess I should call Microsoft to get their opinion." But who do they call for Open Source to get the other side of the story? This is the big question.

Do we want them to call the FSF? Talk to Stallman? Personally, I don't think so. Their religious fervor might well scare off the staffer and/or reinforce MS's position. Perhaps they should call RedHat. Or O'Reilly. But a) I highly doubt that they've heard of RH, and b) why on earth would they call a book publisher about a software issue?

The problem is that there is no one person to call. This is kind of the nature of open source, actually -- a distributed group of people working together for some common goals. But that nature makes us politically vulnerable. And it may become a problem in the future. It's depressing, but true.

There are other ways to influence the process, but it all comes down to money. And MS has a whole lot of it. One way or another, there needs to be some central group(s) organized in the open source movement to start handling the politics of it. One suggestion that was floated was to get the Big Companies who use open source involved (Sun, IBM, Compaq, HP, etc.). Indeed, IBM is going to dump $1B into Open Source this upcoming year. So likely the NY senators and congressmen will be on our side. But that isn't nearly enough. What about all the countless other companies out there that use open source? This sounds crass, but forget all those millions of teenagers and spare-time developers out there who are writing open source, what about the companies who are using open source to increase their bottom line and increase shareholder value? I would think that they would be pretty peeved if open source went away.

MS made the mistake of ignoring the politics a few years ago, and has since taken steps to rectify the situation. We need to do the same, lest we get the equivalent of an anti-trust/monopoly lawsuit against open source. Our BOF leader said that this was the first time he had spoken out about these kinds of topics; we encouraged him to spread the word since he was so well informed.

This was very enlightening to Andy and me. Depressing, but forewarned is forearmed. I don't see either Andy or me getting involved on this front, but we'll see how it plays out.

It was late after this (after 10pm), so we went back to our hotel after this. We went to the keynote the next morning, but didn't pay much attention (it was mainly three people who stood up and said, "we use open source in our company, and it's been mostly great, but we did have to work around some problems."), and just answered e-mail and the like. Then we got on planes and headed home.

All in all, like I said at the very top, it was an interesting conference. If nothing else, the enthusiasm for code was quite contagious. We didn't attend many of the "technical" sessions (there were oodles of meetings on MySQL, PHP, Perl, Python, and all kinds of other open source projects), but the buzz was everywhere.

Go forth and write Righteous Code.

July 29, 2001

I hope you bring your appetite. I'm serving cookies shaped like famous Communist figures!

How many times have you gotten the SirCam worm?

I think I've gotten it at least a dozen times, twice of which were in spanish.

I was just notified by the Army that I go before the promotion board for Captain in November. How funny is that?

I am immensely amused. I can't imagine that I'll get it; I'm quite sure that there are many other 1LT's ahead of me who are more deserving of a CPT slot. But I still find it damn funny.


Playing with new autoconf (ver 2.52), automake (ver 1.5-p4), and libtool (ver 1.4). Here's some things that I've learned:

  • It's not too painful to move to the new autoconf. There's a few macros that have to be changed (e.g., AC_INIT has a new arg list), and some of the macros that I use frequently have been deprecated in favor of new ones, but the transformations are mostly straightforward.

  • There's some new handy fortran macros that will help in LAM/MPI.

  • Remember how the line you invoked configure with used to be in config.status? It's not there anymore. Doh!

  • Not to worry, though, the line that you used to invoke configure with is now in config.log (I don't know why they moved it). In general, config.log is now much easier to read, and has lots more information in it that will be valuable to both sysadmins and programmers.

  • "configure --help" has lots more information in it.

  • AC_OUTPUT is now effectively broken into multiple macros; AC_OUTPUT itself just triggers finishing the write of config.status and then runs it. AC_CONFIG_FILES, AC_CONFIG_HEADERS, AC_CONFIG_COMMANDS, and AC_CONFIG_LINKS are now how you specify the output files, output header files, commands to run in and around output time, and sym links to make.

  • "acconfig.h" is now obsolete. There's a handful of new "AH_" autoheader macros that you put in configure.in to put in the top and bottom portions of the header file. You also have to specify "templates" for each #define with AH_TEMPLATE. I'm not sure how I feel about that one. :-\

  • Some cool new macros that will rapidly become my favorites:

    • AC_ARG_VAR: Mark a shell variable as "precious", list it in the output of "configure --help" and save its value in config.status. Warn if the value stored in config.status doesn't match the present value (in case you run config.status at a later date).

    • AC_HELP_STRING: Automatically format the strings that you give to AC_ARG_WITH and AC_ARG_ENABLE to get that pesky spacing right.

    • AC_SUBST_FILE: Allows you to do the same thing as AC_SUBST, but substitute in the contents of a file. I see
      $COPYRIGHT$ potential here...

    • AC_CHECK_DECL: Checks to see if a symbol is declared. I've always had my own tests for this; I'll be happy to start using this macro instead of my own.

    • AC_CHECK_MEMBER: Check whether a struct or class has a given member or not.

All in all, it looks like they tried to add significant useful functionality to autoconf, so I'm overall pleased.

HOWEVER, it caused me an hour of two of frustration before I finally tracked down a problem with libtool -- there's a bug in the libtool 1.4 distribution. If you use AC_CONFIG_AUX_DIR to put all your config files in a subdirectory rather than the top-level directory (which I do), and if you use configure.ac instead of configure.in, libtool will get corn-fuzed and put ltmain.sh, config.guess, and config.sub in the top-level directory instead of your config directory. This eventually causes much badness... no need to discuss specifics here.

The problem is in the "libtoolize" script -- there's one place where "configure.in" is still hard-coded in, instead of using "$configure_ac", which is set to the Right value.

This has apparently been fixed in the libtool CVS (I checked), so the next version will have this correct. Ugh.

That's enough for now. Time to go mow the lawn.

July 31, 2001

I ate, drank, and slept Tap.

I found an old journal entry that I never finished... Since Telocity is having connectivity problems (first it was nd.edu this morning. Now Telocity. I was having shaky connections this morning, and now I can't connect to anything. Ugh.), I might as well finish this off even though the events are from about a week or two ago.

Random factoid that I learned today -- the first digit of IP addresses indicates whether the address was a class A, B, or C.

ND football lottery: Won Pitt, lost all others. Oh well; I'll still be a student at ND this upcoming fall, so I'll still get married student tix.

I'll see Rich tomorrow at the Open Source Conference.

I forgot to plug D's absurdly cool house-o-meter. http://www.kresge.com/hstat.html.
Tracy and I spent the day at lake w/ Janna and their new boat water skiing and tubing. Janna are good at skiing. I apparently had some spectacular "dismounts" from the tubing, including some uncontrolled cartwheels across the water. I somewhat hurt my left shoulder/bicep when I went over some huge bump (got airborne, too) on the tube. I also made the mistake of taking my shirt off to go swimming in the lake, and totally forgot to put sun tan lotion on my shoulders. Doh.

I fixed my car's stereo problem. The wiring was causing screeches -- whenever I went over a big bump, the exposed wires would brush up against the metal speaker casing, causing the screech. Funny, though, it didn't look like it had ever had insulation.

We got our tax rebate today -- I have to admit that I was a bit surprised. But hey -- cash is cash. (I've since deposited the check and it didn't bounce -- yay for the US government!)

Good quote from Brian: "Sandia: Summer camp with nukes."

Dave, did you mom change her phone number again?

I heard on the radio that England's Queen Mum's birthday is this Saturday.

She'll be 101.

While driving to Bloomies yesterday, I heard Kevin Bacon and his brother (Michael?) sing some songs from their current album. They're surprisingly good musicians. They had some funny stories about their pasts and Kevin's movie career, too.

Lummy has been very pleased with his Linksys router that he uses to gateway his DSL traffic to his FAN (family area network). It doesn't show up at all to an nmap scan, but he can directly ssh to one of his internal boxen. It seems to be a fairly powerful firewall, allowing a variety of configurations including IP forwarding to up to 10 ports. Check it out.
So I bought one myself; it should be here in about a week. They have models with embedded 4 and 8 port switches, but I already have an 8 port switch, so I just got the single port version of this router. It will certainly be easier than maintaining that dual-NIC box that I have right now, and be [hopefully] more secure, on multiple levels.

I've received the SirCam virus from the same idiot (someone at @optonline.com) about 10 times so far. I've replied each time saying, "Update your anti-virus!", and I have started CC'ing the postmaster. A bunch of my replies have come back "Over quota". Idiot.

There's still a number of bugs left in OSCAR 1.1; I don't think we're going to make a release by Wednesday. :-(

It looks like the router in the engineering building at Notre Dame is fried somehow (engr-e06.gw.nd.edu) -- no traffic is able to get through it. This means that I can't get to my e-mail.


About July 2001

This page contains all entries posted to JeffJournal in July 2001. They are listed from oldest to newest.

June 2001 is the previous archive.

August 2001 is the next archive.

Many more can be found on the main index page or by looking through the archives.

Powered by
Movable Type 3.34