NYCBSDCon 2010: notes


My NYCBSDCon 2010 summary, or How I Spent My New York City vacation:

Things I learned on Friday: The only event was a get-together at the B Bar and Grill, down the street from the facilities.  I realized I could recognize people better if they had their email address on their shirt.

Things I learned on Saturday:

  • The BSD Certification group is a non-profit, and could use donations.
  • What they are doing is a good idea, and it’s very scientifically constructed.
  • Psychometrician, the specialist who helps design the tests, is a great job title.
  • The Automated Test Framework would be great for DragonFly, too.  The next generation test was mentioned.
  • My takeaway from the session on radmind/LDAP for administration was “Use puppet.”
  • Calyptix has managed to build OpenBSD machines into almost a managed service.
  • They have tcpdump on a web page from the appliance they sell.  That isn’t necessarily that hard, but it makes an incredibly powerful tool very accessible to more people.
  • PCSYSINSTALL sounds like a good replacement for SYSINSTALL in FreeBSD.  Frankly, a poke in the eye would be a good replacement for SYSINSTALL, so the bar is low.
  • Jeremy Reed is writing a colossal history of BSD, with so much detail pulled up already he couldn’t even summarize it all in his hour of speaking.  It’s a huge project, with some very entertaining anecdotes.  It also shows the issues we face nowadays are not really new.
  • An early worker on BSD virtual memory was Özalp Babaoglu, which is the best name I’ve had to type yet.  (I’m missing a ~ over the g, too.)
  • Early BSD licenses were surprisingly restrictive, tying software to particular hardware.  Back then, 3BSD cost $200, and the equivalent version of Bell Unix was $28,000.  You can guess what most people bought.
  • sendmail was originally named “delivermail”.
  • Jeremy Reed has a separate book publishing operation.  He’s selling pfSense: The Definitive Guide, which went like the proverbial hotcakes.  (I bought one.)
  • The IPv6 talk was … depressing.  The talk was good, but the evaluation of IPv6 adoption isn’t pleasant news.  People just aren’t switching fast enough, and vendor implementations are generally poor.
  • Jason Dixon did a “Choose Your Own Adventure” presentation, which hopefully will be online soon.
  • The evening concluded at the B Bar again, where I got to have some good conversations with Dru Lavigne, Michael Lucas, George Rosamond, and others.

Things I Learned on Sunday:

  • Matthew and I had issues with the wireless service available at the con all through Saturday.  Sunday morning, Matt turned around and said “here’s the fix” – a few lines turning the hardware encryption off.  Several of the various laptops at the convention using Atheros wireless chips had the problem, under other BSDs.
  • In addition, terminal beeps can now be turned off, cause it was just annoying.
  • Isilon (which was just bought for billions of dollars, holy crap) had a presentation on their FreeBSD-based multi-node storage product.  It sounds interesting; they use SSDs to cache metadata.  I’ve heard of that strategy before…
  • James K. Lowden had a presentation that took the form of a call to arms; it can be summed up in a quote:

    [SQL] is the biggest most popular worst language in the world.

  • He had some very good points about how databases all use SQL and all are totally incompatible in terms of tools and data interchangeability.  This was interesting enough I missed the session for the various BSDs to report status.  The Third Manifesto talks about a language called “D” as an alternative.  freedb.schemamania.org is another place to look – the slides are linked there, and the history page is a good summary.
  • James Lowden is responsible in part for FreeTDS, which I’ve found very useful in the past for connections from Unix-land to Microsoft SQL Server.  He was also the snappiest dresser at the convention.  This was the most energizing talk.
  • George Rosamond gave an informal talk about the state of the con.  This was the best year yet in terms on attendance and sessions and facilities, and it looks like the con is going to come out with positive cash flow.  All good news.
  • The pfSense presentation made me buy a book, as I mentioned before.  This will be useful for work; the pfSense product is very full-featured.
  • It sounds like the changes needed for pf, in pfSense, will make it back to FreeBSD 9.  Some of these changes would be great to have in DragonFly.  Apparently some sort of dummynet is being implemented, which may mean ipfw2 could be dropped.
  • Michael Lucas’s “BSD Needs Books” was funny.  The two good quotes I wrote down were:

    “Someone should do this.  Someone is you.”

  • and, when he was asking for book title suggestions from the audience, someone suggested “Kernel Hacking for Complete Idiots”.  His reply: “That would be a Linux book.”

Summary: a great convention.  The facilities were great, the catered food was excellent, and the topics were fun.  The weather even cooperated.

In the future, watch for Jeremy Reed’s Unix history book; it will be more than a year, probably, before it hits the shelves, but it will be important.   Also, if you’re going to listen to my prescriptions, get certified when the chance arises.

Posted by     Categories: BSD, Conventions, Goings-on, UNIXish     15 Comments
15 Comments on NYCBSDCon 2010: notes

Respond | Trackback

  1. Ivan says:

    Really nice! Were any of the presentations recorded?
    Would be nice to have them online.

    And Michael Lucas’s quotes made me laugh hard hahaha

  2. Jason Dixon says:

    How did we not run into each other at the conference? :(

  3. Regarding your comment on 3BSD pricing … they still had to buy the expensive license first :(

  4. Ha ha! I feel like such a celebrity! I was the idiot that suggested “Kernel Hacking for Complete Idiots”… :-D

    And Jason, your nethack was def the high point of the Con for me! Are you planning to release it? :-)

  5. Jeremy – I remember you mentioning that now. So they purchased Unix, and then spent $200 more to make it work? :)

  6. Jason, after your presentation I figured out who you were, but never had a point where you were free and I was free at the same time. See, now if there was a DCBSDCon, I’d totally be able to find you.

  7. Ivan – the presentations are supposed to show up on the NYCBSDCon site. There was video recording being done, so slides with audio should be available, though I’m sure it’ll take a bit to get the hours of footage together.

  8. Will Backman says:

    I had to leave early on Sunday, but here are the recordings of the ones that I could get:
    http://cisx1.uma.Maine.edu/~wbackman/bsdtalk/NYCBSDCon2010

  9. blinkkin says:

    Thanks Will! Audio is available (at least for few talks) , now where I can find papers?

  10. Lawrence Teo says:

    Thank you for posting this, Justin! I think it is a great summary of the con.

    Siraaj: Jason’s “nethack” is available at http://dixongroup.net/cyoa/ .. and yes I still owe you slides :)

  11. mwlucas says:

    Siraaj, you’re not an idiot. Someone had to provide the straight line. :-)

  12. Joe "Floid" Kanowitz says:

    re: IPv6 – is there a favored dead-tree guide that’s actually been updated since 2005, or what’s everyone using to catch up on the more esoteric / ‘dangerously different’ aspects?

  13. If there is an IPv6 guidebook that is up to date, I’d like to see it.

    My current IPv6 implementation/migration plan at work consists of looking in my server room and sighing.

  14. […] took place on 14 Nov. Justin Sherill has summarised the 2 days: NYCBSDCon Notes. Will Backman, the man behind bsdtalk, has uploaded a short video showing off the facilities, food […]

  15. […] Like I said for the last NYCBSDCon in 2010, it’s totally worth going.  I now have a long, long list of things I want to do and ideas to try, all from meeting people face to face and talking about what we can do.  It’s energizing, far more than meeting over IRC.  A third of the people there had no prior BSD experience.  George Rosamond mentioned that he was thinking they could do this perhaps every 6 months. […]

Respond

Comments

Comments