27 posts tagged “squeak”
Next week at OSCON 2009 in San Jose is going to be a very busy week for me. I'm managing the Stonehenge/LinuxFund party on Wednesday night, which includes hanging out at the LinuxFund booth during the day on Wednesday. I'm also appearing at the usual O'Reilly Author meet-and-greet Wednesday evening.
I had also been accepted some months ago to teach a paid 4-hour tutorial "Introduction to Web Application Development Using Smalltalk Seaside" on Monday afternoon, and a 45-minute talk "Persistence Solutions with the Smalltalk Seaside Web Framework" on Thursday morning. For some reason (which I'll blame on the drama around my broken arm), I forgot to blog about those here, although I'm sure I Twittered about it. :)
But on top of that, I had originally pitched an "intro to smalltalk" course that would precede my Seaside talk to prepare people for the afternoon talk. It had been summarily rejected, and now I know why. Avi Bryant (co-creator of Seaside) had been given that task Monday morning. But he wrote me a couple of days ago, asking me to take the talk over, as apparently he will be unable to make it to OSCON. So, now I'm also doing roughly the talk I had originally pitched as well: "Finding the Swan in Squeak's Ugly Duckling" on Monday morning. And I'm now scrambling to prep for that too, although a lot of it had already been planned for my 5-day "intro to Smalltalk and Seaside" course that Stonehenge will be offering soon.
So, there was one talk on Smalltalk last year at OSCON, and now three talks this year (including two paid tutorials). I'd say Smalltalk is making a good comeback.
As I just posted to squeak-dev:
In trying to recap this past year, my first year of being back into the Smalltalk world, I googled "randal schwartz squeak", and saw:
Results 1 - 10 of about 5,910 for (randal OR randall) schwartz squeak.
I think about 5700 of those pages didn't exist a year ago. Here's the highlights:
- Created my blog methodsandmessages.vox.com to publish squeak news and observations
- Demonstrated Squeak and Scratch on the "Lab with Leo" show, hosted by Leo Laporte and distributed in Canada and Australia
- Had both Avi Bryant and Dan Ingalls on FLOSS Weekly, my weekly podcast with an audience of about 25,000 or so. Squeak was mentioned once or twice. :)
- Managed to sneak a mention of Smalltalk and Squeak into just about every other episode of FLOSS Weekly
- Got Leo Laporte thinking about a weekly Seaside tutorial video podcast, in which we would run through my courseware to teach him (and the audience, by proxy) how Seaside and Squeak works
- Provided the weekly Squeak News segment for James Robertson's Industry Misinterpretations podcast, broading the continued presence and currency of Squeak throughout the Smalltalk community
- In my role as Leadership Team member, coordinated the cutover of the squeak.org to Aida, and am currently sheparding the Squeak 4.0 relicense effort
- Established the meme "the year of Smalltalk" in my blog, which has been used as a unifying rallying cry to help the resurgence in 2008
- Got the Brazillian FISL conference interested enough in Smalltalk from my 2008 presentation that they created an entire Smalltalk/Squeak miniconf for 2009 (which I'm just now putting out the CFP for, by the way)
- Got interviewed by InfoQ about the Smalltalk Comeback, where I mentioned Squeak prominently
- Got interviewed by WebDevRadio about Seaside, mentioning Squeak at least a few times
- Got interviewed by Reflective Surface, a blog about Ruby-ish things about Seaside and Squeak
- Produced a keynote talk at STIC's Smalltalk Solutions 2008, which included my Seaside talk and a Persistence talk, both incorporating many mentions of Squeak
- Gave a three-hour lecture at OSCON 2008 on Smalltalk and Seaside, using Squeak as the primary demonstration tool
- Produced a few videos for Vimeo on how to use Squeak, and got a few other people using Vimeo for hosting, including creating a Squeak group there
- Created a "Seaside Bootcamp" class, to be delivered by BigNerdRanch in Germany in a few months
- Promoted Squeak via watching various twitter and blog messages for mentions of Smalltalk, Seaside, and Squeak
- Answered many questions on squeak-dev, beginners, and seaside mailing lists.
- Hung out in the #squeak, #seaside, and ##smalltalk channels on freenode, and the #smalltalk channel on Parcplace, answering questions and occasionally annoying the other locals
- Gave the "Seaside: your next web framework" advocacy presentation at many companies and conferences, which features Squeak as one of the many Smalltalk implementations that runs Seaside
- Presented a talk on Squeak GLORP at the PostgreSQL day in Portland
- Got interviewed by GeekCred podcast for a general biography, but managed to get Squeak mentioned in there once or twice
- Answered Smalltalk and Squeak-related questions on Stack Overflow
Ok, I'll stop there. I'm sure there's three important things I'm leaving out, and 20 trivial things.
Looking back at this list, all I can say is... "oops, I set the bar too high for my second year". :)
But seriously, I have dedicated, and continue to dedicate, my time and energy and attention and skills to promoting Squeak as the basis for a sane programming environment to make it easier for all of us to get things done, and cooperate and collaborate in an open-source way to avoid repeating ourselves.
I would be honored to serve a second term on the Leadership Team, to help complete the relicensing effort, and work with all parties to figure out what Squeak 5.0 needs to be. We're at an interesting point in Squeak's development (think: "the teenage years"), and continuity with the past and a vision for the future are cruicial.
Please consider me as you cast your ballot.
The InfoQ website interviewed me recently about Smalltalk and Seaside and Squeak. It's a nice article with a lot of external references. Check it out!
I keep mentioning Squeak Smalltalk as I'm chatting with Leo Laporte while taping my well received podcast FLOSS Weekly. It's actually quite fun. He keeps bringing it up as something he's learning, and I think it's giving a big visibility to Squeak and to Smalltalk in general. There's even talk of creating a special Learning Squeak series that would be shown on Leo's live webcast network. Check it out!
As I was reading the latest entry in Brett Hallett's blog, titled "Rewriting a script into a callable method", I saw the code:
i := 20.
d := '0123456789abcdefghijklmnopqrstuvwxyz'.
l := d size.
s := String new: i.
1 to: i do: [ :x |
s at: x put: (d at: (Random between: 1 and: l))
].
s printNl.
This code didn't look very smalltalk-like to me. So I thought a bit about what was striking me odd, and it was all those explicit #at: and #at:put: calls. Not to mention confusing uses of "i" and "l" in a bad font. :)
The first two lines aren't bad. I'd keep them, although I'd probably name the values a bit more intention-revealing:
desiredLength := 20.
sourceCharacterSet := '0123456789abcdefghijklmnopqrstuvwxyz'.
sourceCharacterSet := '0123456789abcdefghijklmnopqrstuvwxyz'.
But the rest of the code starts going a bit odd. Consider the expression:
d at: (Random between: 1 and: (d size))
(factoring out the temp variable). This is equivalent to simply:
d atRandom
And the loop there is collecting together "desiredLength" number of these random selections. So let's say that using a Collection-style method instead:
(1 to: desiredLength) collect: [:each | sourceCharacterSet atRandom]
We're now much closer to the result. The problem is that we now have an OrderedCollection of Characters, not a string. But a simple class-mapping will fix that, because Strings know how to create themselves from Collections of Characters:
((1 to: desiredLength) collect: [:each | sourceCharacterSet atRandom]) as: String
And there we have it. If #collect: didn't barf on a 0-arg block, we wouldn't need the misleading ":each" in there either. (Developed with Squeak... not sure how unstandard this stuff is, or if it's just missing in GNU Smalltalk.)
Wow. What a rush. But I mean that literally.
I've known for a few months that I'd be delivering a three hour tutorial on Seaside at OSCON. But like many of the activities, I didn't actually start writing the materials until last week, and I'd been sweating bullets trying to generate what I believed to be enough material to fill three hours.
I'd settled on a basic outline that included:
- A quick intro to Smalltalk and the Squeak GUI
- Basic concepts of Seaside (starting from configuring a "hello world" application)
- Forms and state
- Subcomponents and call/answer
- Magritte
- Persistence
- Other bolt-ons (testing, ajax, etc.)
I had decided to use a "demo first, then summarize what I showed on slides" approach, because this means that the slides could be a bit more outline-like, since they didn't need to actually teach the material. This is a different teaching style from what I am most familiar with—the Perl classes are almost all slide-driven, not demo-driven.
I decided for the demo that I would create and elaborate on a simple "To Do List". Yes, I was inspired by the now-classic Potsdam Tutorial, but I came at it my own way, trying to figure out how to change the fewest lines of code to get to the next working program and yet continually illustrate a new feature each time. I rehearsed my presentation Saturday with Tom Phoenix, my fellow Stonehenge presenter, who was also going to be in the front of the room with me in case I passed out or got tongue tied. When I realized that this particular example really illustrated a lot of good ideas, I knew I was on the right track.
When I had first pitched the idea to OSCON, I wasn't sure what kind of turnout I'd have. After all, my Seaside talks hadn't been accepted to the main OSCON tracks (I believe they were deemed "not relevant enough"). So with the idea of an additional three hours for which people would have to pay to attend, and on Monday (the most distant day from the rest of the conference), I had decided that I would be "extremely pleased" if as many as 15 people showed up.
At Smalltalk Solutions 2008, James Robertson of Cincom offered me the opportunity to hand out Cincom VisualWorks installation disks to my students, which I readily accepted. I said I'd get back to him with a number so he could ship them to me. When I asked O'Reilly what my pre-registered number was, I nearly fell out of my chair when the response was a whopping 41 people. James sent me 50 disks (bizarrely each individually bubble wrapped, and in a box that also had bubble wrap filling), and that was followed by Monty Williams of GemStone quickly burning 50 DVDs of GLASS for me to hand out as well.
And, it turns out that by noon today, a total of 55 people had signed up! The room was nearly packed. The people were attentive, and asked good questions, and were generally following along, including being in wonder and awe when I got to the intra-hit debugging and talked about how the continuations provided a level of control-flow abstraction.
But, here's the crazy part. By the midway break, I'd covered only the basics of Smalltalk. I hadn't even started the Seaside part! I spent the last 90 minutes building the basic "To Do" application as a simple single-page form, adding items from a field at the bottom, and having a "delete" button by each item to remove it. So I had clearly underestimated the amount of time it takes to walk through the material, and had clearly over-prepared.
I didn't even get into changing each item from a simple string into an Action object, creating a subcomponent for the bottom form, or using call/answer for validation. None of the stuff that I had prepared beyond the initial code. Well, sounds like I really have eight hours of material already. That's good, because I'm going to need it on my next delivery. Yeay.
In short, my first "big" Seaside tutorial went very well, and the feedback was great, and I suspect quite a few of the students will be experimenting with Seaside (or at least talking about it), if not strongly considering it for a near future project. Smalltalk and Seaside clearly are on the radar of many people now, and the attendance and the attentiveness of this tutorial shows that we are on the right plan.
But I'm beat. And tomorrow's another day.
Sunday, I spoke at the pre-OSCON Portland Oregon PostgreSQL Users Group (PDXPUG) DAY mini-conference about how to connect Squeak to Postgres using GLORP. I had originally pitched the talk with about 45 minutes or so of material in my head. When I got the final schedule, my time had been cut back to 20 minutes (and the coordinator wanted half of that to be "introduction to Smalltalk"), so I realized I could really only do a simple demo. Of course, I was working hard on my materials for my 3-hour OSCON tutorial the following day, so I waited until Sunday morning to actually start writing my PDXPUG talk.
I created the classic BlogPost and BlogComment tables a little bit at a time during my testing. I also set up a Workspace with a series of Do-Its and Print-Its that I knew I could simply run down and fire off to show what was happening. Once I had finished that, I created another workspace that showed a series of Print-Its to show off the basic Smalltalk syntax. I then resized my image to fit into the 1024x768 projector size, and saved it, and waited.
Well, it turns out that lunch ran an hour late, and since my talk was in the afternoon, they asked me to shorten it up a bit. In 5 minutes, I quickly ran through all of my Smalltalk Print-Its. In another few minutes, I showed off the table definitions and object descriptions of my GLORP system, and then I ran through all of the Do-Its and Print-Its of my GLORP demo. Total time: about 12 minutes.
When I sat down (without time for questions), I was sure that it had been a waste of everyone's time. But much to my surprise, the coordinators came over and said "great job", and one-by-one as the afternoon progressed, many of the attendees came up to me asking specific questions that definitely related to my presentation.
So, don't be dismayed if you get only five or ten minutes on a longer agenda somewhere. Squeeze in some Smalltalk. There's always room.
Michael Kimsal runs a nice podcast on various modern web development topics, and agreed to interview me recently about Seaside. The show is now live. It's basically a more detailed version of my "Seaside: Your Next Web Framework" talk but with some side conversations thrown in, including my passion for Smalltalk and Squeak, and even a bit about MagLev. Check it out!
My friends at MySpace.com finally transcoded my recent presentation at their headquarters of yet another version of "Seaside: Your next web framework", including a rather badly botched demo of Squeak. I haven't watched it yet, but feel free to review it for yourself. Please note that, yes, I made a few technical errors and have been amply corrected. Enjoy.
James Robertson of the weekly Smalltalk Industry Misinterpretations podcast has graciously offered me the opportunity to add a short weekly Squeak News segment to his podcast. I'll be selecting the most recent posts from news.squeak.org, and reading them at varying speeds, depending on the amount of news. Hear the result at the current podcast, and subscribe to get future (weekly) releases.
It would be really cool if my appearance reading the Squeak News gets picked up for Squeak News, so I'll be reading an article from Squeak News for the podcast about reading the articles from Squeak News for the podcast. Turtles all the way down!
