Antony Blakey has some interesting ideas about making Smalltalk more accessible to the non-Smalltalk developer.
Chris Pirillo ran across an all too common problem in the IT sector - lousy support, with Comcast being the louse of the day. I can relate to his experience - I've had many, many encounters like this with Comcast:
My connection kept going up and down, but by the time I was able to speak with someone in the support department, everything was fine again.
“Well, sir, we’re not showing any problems with your connection right now.” Yes, it’s fine right now - but not when I try to open up more than one Internet app. I was even having issues with Vonage and Skype… “Well, sir, we can’t be responsible for programs that we do not officially supp…”
Whoa, whoa, whoa - WAIT A MINUTE! I interrupted him, loudly. I’ve been using Skype and Vonage and streaming live video and doing all sorts of wacky things across my home network and never had an issue I wasn’t expecting to experience. Vonage wasn’t the problem. Skype wasn’t the problem. My browser wasn’t the problem. The Netgear gateway router that Comcast gave me was the problem.
If I had to guess, the problem goes right back to the motivations of the support staff. They are probably rewarded based on how quickly they handle calls (or how many they handle in an interval of time, which amounts to the same thing). The rationale for such schemes is to save money - although, if you sit down and consider that for a few minutes, the rationale doesn't hold up well in an era where an arbitrary user can end up being an accidental PR nightmare.
In any event, the "get them off the phone quickly" theory holds up well in the face of experience with Comcast. If you have anything other than Windows connected to the Comcast provided router/cable modem, the phone tech will try the "we don't support that" dodge immediately. The fact that the client is irrelevant in most cases doesn't enter into it - it's a dodge to get rid of you.
I've said before that it's a bad idea to let lawyers become ad-hoc PR staff (lawsuits). It's even worse to let your support staff cause PR headaches due to poorly thought out compensation plans. Somewhere at Comcast central, a bean counter thinks his support plan is saving the company money. What he doesn't realize is that the PR staff down the hall wants to know who keeps hitting them with the ball peen hammer...
Technorati Tags: tech support
On today's Smalltalk Daily, we continue with the syndication example, adding filtering to it.
I guess you could extend my title - just don't let your lawyer speak in public, period. Here's Sony BMG's anti-piracy lawyer, calling many of Sony's customers thieves (via Ars Technica):
Pariser has a very broad definition of "stealing." When questioned by Richard Gabriel, lead counsel for the record labels, Pariser suggested that what millions of music fans do is actually theft. The dirty deed? Ripping your own CDs or downloading songs you already own.
Gabriel asked if it was wrong for consumers to make copies of music which they have purchased, even just one copy. Pariser replied, "When an individual makes a copy of a song for himself, I suppose we can say he stole a song." Making "a copy" of a purchased song is just "a nice way of saying 'steals just one copy'," she said.
Mind you, this group of individuals includes people who own Sony hardware - various MP3 players, the PSP, the Vaio - the list goes on a long way. Is Pariser consulting with Sony's hardware division so that they can build in heavy-handed ways of preventing this kind of *cough* theft *cough*?
I love hot food - and I love Thai food - so I really, really would love to eat at this place:
A Thai restaurant has been cordoned off by police after its extra-hot homemade chilli sauce was mistaken for a chemical outbreak.
Firefighters dressed in special suits smashed down the doors to discover the source of the smell - chef Chalemchai Tangjariyapoon's fiery signature nam prik pao chilli sauce.
Sadly, London is a long ways off from here :)
In today's "how stupid can the music industry be?" segment, we learn that their anti-piracy campaign is not only stupid from a PR standpoint - but it's a money pit without clear objectives to boot! This is the kind of plan that only a lawyer who's paid by the hour could love - via Ars Technica::
The next line of questioning was how many suits the RIAA has filed so far. Pariser estimated the number at a "few thousand." "More like 20,000," suggested Toder. "That's probably an overstatement," Pariser replied. She then made perhaps the most startling comment of the day. Saying that the record labels have spent "millions" on the lawsuits, she then said that "we've lost money on this program."
The RIAA's settlement amounts are typically in the neighborhood of $3,000-$4,000 for those who settle once they receive a letter from the music industry. On the other side of the balance sheet is the amount of money paid to SafeNet (formerly MediaSentry) to conduct its investigations, and the cash spent on the RIAA's legal team and on local counsel to help with the various cases. As Pariser admitted under oath today, the entire campaign is a money pit.
The stupidity on display here is just stunning. I wonder if there's anything resembling a clue over at
stupid central the RIAA.
I've stayed away from the "iPhone bricking" story, and it looks like it was a good idea not to draw too many conclusions. The entire pundit class (see Nick Carr's post on it for an example of the genre) was up in arms about how "evil Apple" was bricking phones in some kind of revenge act.
The iPhone bricking problem has been a PR disaster for Apple, making the company look punitive and obsessed with control. But Erica Sadun, a technical writer and blogger at TUAW.com who contributed to an iPhone unlocking application, said Apple's update wasn't designed to disable hacked devices. Just the opposite: Sadun thinks Apple worked hard not to brick iPhones -- even hacked ones.
The new iPhone software appears to be a ground-up rewrite, unrecognizable under the hood to the older version, which Sadun said was "very unfinished" and, in some places, "a complete hack."
Hmm - where have we heard this story before? Version one is rushed out the door, followed by a quick patch that looks a lot like a rewrite. Only for every second software product. Look at Twitter and its famous outage issues in the early days of their initial traffic spike - they were madly trying to replace the parts of their application that didn't work out. Looks like Apple did the same thing, only more quietly. The reality is, Apple has always desired stability over hackability - they are targeting consumers, not geeks.
Some people look at the easy hacks available for Windows and see nirvana. Apple looks at the same thing and sees hell.
I don't know that the .NET "shared source" opening is a trap for Mono, but I do know this: the fact that it is being considered one by some writers demonstrates the depths of mistrust Microsoft has managed to set up around itself. Consider:
Let's say a year from now, Microsoft does a SCO. They claim that Mono contains code that was stolen from the .NET Framework reference source code. They point at their code, they point at the license, and sure enough, there's similar code. After all, both projects are implementing .NET; there will almost certainly be lines of code that looks alike.
Better still, from Microsoft's point of view, all they need to do is find one Mono programmer who has signed the license to look at the .NET Framework reference source code. With that "proof," they'll claim they've found their smoking gun. SCO failed in its attempts because it never did have any evidence that there was Unix code in Linux.
Microsoft, however, is baiting its trap for Mono programmers with .NET cheese. They'll claim, come that day, about how open it was in letting people look, but not touch, their code. With the combination of "proof" that some Mono code has been stolen from Microsoft and its attempt to muddy the waters about what open source really means, it can look forward to having a much better chance of killing off an open-source project than SCO ever had with Linux
I think any such attack on Mono is unlikely, and that it would be a PR disaster if they did make that kind of attack. But never mind - the idea is being rationally discussed. That's a loss for Microsoft right there.
Technorati Tags: PR
Michael tipped me off to this last week - last night he also told me that I had to enable DirectX in the Windows VM. After doing that:
Last week I upgraded Parallels v3.0 to the latest build. For whatever reason I noticed that I could now play Subspace under Parallels where before I couldn't. Puzzled by the sudden new capability - I fired up Civ4 Beyond the Sword to quench my curiosity. To my surprise, it runs!!
I played a game until past my bedtime to prove it to myself last night :)
This is just classic. A California government agency ran into trouble last week - they reset some domain server information after one of their systems got hacked, and the domain ended up under the purview of the GSA. That was a problem, because the issue wasn't resolved fast enough, and the GSA's admins apparently forgot that California has more than one website:
On Tuesday, at around 2 p.m., the federal organization responsible for managing the .gov top level domain pulled the plug on the ca.gov domain, according to Jim Hanacek, a public information officer with California's Department of Technology Services. The "ca.gov domain was removed as a valid address by the federal General Services Administration, who has an office that oversees the use of the .gov domain," he said.
Fortunately for California, they noticed and got on the horn with GSA. Had they not noticed right away, and DNS records propagated, there would have been chaos. I'd love to know who the bright guy at the GSA was who thought this was a good idea.
Update: Phil Windley explains that the Feds went way, way too far - both from a common sense standpoint, and from a legal standpoint.
Technorati Tags: security
All I had was my phone, but I liked the way this web looked in the morning fog, at the base of the lamp in my front yard:
It was intriguing for a day or two, but now it's clear that the Leaderboard was the dumbest idea ever, because now more than ever, people are gaming Techmeme so they can climb the list.
To which I'd say, "well, duh". And if Techmeme didn't make that kind of list, someone else would. People like lists, on the web and off it. There are many things are that just beyond Winer's ken; this is simply one of them.
The folks at Precision have sent me a list of open Smalltalk positions - here it is:
Precision Systems (www.precisionsystems.com) currently has Smalltalk positions open all across North America, from Los Angeles to New York and everywhere in between! Please contact Vicki Ross (recruitVR@precisionsystems.com) if you’re interested in any of the following positions:
Waterbury, CT – 1 position
Smalltalk Trainer, 2-3 month contract
Minneapolis, MN – 2 positions
Smalltalk Developers, contract and contract-to-hire
Northern New Jersey – multiple positions
Senior Smalltalk Developer, permanent, 6 month contract-to-hire, and 12+ month contract
New York, NY – multiple positions
Smalltalk Developer, Smalltalk/Java Developer, contract and permanent
Texas – 2 positions
Smalltalk Developers, 6 month contract, AND long-term contract or 6 month contract-to-hire
Don’t forget to inform your co-workers and friends; for any new and successful referral to Precision, we will pay you $1,000!
Technorati Tags: smalltalk
Mathew Ingram has pegged the music industry:
I’m not going to argue that what Jammie Thomas did was right in a legal sense, because it clearly wasn’t, as Mike Masnick notes at Techdirt. So the RIAA was obviously within its rights to sue. But $222,000 for 24 songs? That’s just ridiculous. It’s a good thing the case only involved 24 songs, and not the 1,700 or so that Ms. Thomas had on her hard drive initially. That would have left her paying about $15-million for that music collection, if the same formula was used. And what was the formula? Something like X times Y, to the power of Z — where X is the lack of a sustainable business model, Y is an aggravated response to a non-existent threat, and Z is the inability to differentiate between customers and thieves.
Like him, I agree that this woman should not have grabbed music without paying. However, the penalty is absurd, and all it's going to do is convince the RIAA that their "we hate our customers" jihad is a good idea - which means we'll see more of it. Oh happy day.
I said awhile back that I thought the various legal assaults on Microsoft were pointless - give them enough time, and - like IBM in the 80's - they'll stupid themselves out of their influential position in the industry. It seems that they are ahead of schedule on that front - Joe Wilcox is on the "whither Vista" beat again. He raises a number of anecdotal stories about resistance to Vista, and then makes this observation:
Many IT organizations are intolerant to change that disrupts the workflow. Windows Vista is a disruptive force. Pick a reason: Application incompatibility, hardware incompatibility, UAC popups or user resistance, among others. XP is familiar and, for many, feels safer.
Microsoft should be hugely concerned about the stability of the Windows XP ecosystem and the operating system's customer familiarity. Windows shouldn't become another WordPerfect.
Clean breaks are possible - Apple made one from OS 9 to OS X. However, any clean break has to offer substantial benefits over the old stuff, and OS X did, in terms of stability. The problem for Vista is that XP is "good enough", and the persistent horror stories about drivers and UAC lead people to ask "why should I bother?".
Technorati Tags: Vista
Salon's "Ask the Pilot" guy explains why flights are getting later and later - it's a traffic jam:
Carriers have created this mess through a self-defeating insistence that frequency of flights is the ultimate key to success. Over the past several years, they have portioned capacity onto smaller and smaller planes making more and more departures. The results of this strategy can be seen on any afternoon at airports such as JFK, Newark, LaGuardia and Washington National, where small regional jets (RJs) account for up to half of all takeoffs and landings. It is not the total volume of passengers slowing things down, it's the inefficient way they are divvied up. In some places, 50 percent of the traffic is carrying a quarter of the people.
It's a good article that explains the problem (and why solutions are probably unlikely) pretty well. Air travel has always required patience; it's probably not going to get better anytime soon.
If you've tried loading the Seaside parcels into ObjectStudio 8, you've likely run into problems. However, we have a fix for that, and it involves applying a small patch - here are the steps: (you can click through any image for a larger view)
Caveat: Loading Seaside may well cause problems for existing ObjectStudio code. We will have those issues addressed in a future release, but if you load this into the current ObjectStudio 8, you should not use an existing project image - meaning, you can safely explore, but don't try to deploy a combined Seaside/OS8 application yet.
First, browse to class EncodedStream, and look for the method #skip:
Now, restore the original code via the "Restore Overridden" menu pick:
Now, open a Parcel Manager from the Launcher, and load Seaside for Web Toolkit:
Now follow that url!
That's it - when we ship the next major release, this patch will be part of the system
Well, the Yankees are off to a terrible start in the playoffs. Tonight's game had its bizarre moments, too - Chamberlain was driven to distraction by the bugs that swarmed the field in the 8th, and gave up an unearned run on two wild pitches. Carmona wasn't rattled, and hung in for nine solid innings. Things rolled along tied until Rivera left the game, and that was that. Looks like the post season will be short this year.
David Buck has been doing some very interesting work in Cincom Smalltalk:
I've been doing some interesting work for Cincom that should be appearing in a future release of VisualWorks. I now have the ability to call VisualWorks as a DLL (on Windows) or a shared object library (on Linux or *nix). On Windows, for example, you can use ResHacker to add the image to the VisualWorks DLL and call it from a C application.
Technorati Tags: smalltalk
Dare Obasanjo writes about the Dynamo storage system at Amazon, and how it provides services that an RDBMS really can't. It occurs to me that I should have asked the Gemstone folks about these kinds of issues when I talked to tem a couple of weeks back (that podcast will be out next week). I am no expert in the storage arena, so I won't pretend to know whether an ODBMS is a better answer than an RDBMS at the scale being talked about here - but it would have made a great topic :)
I got an email with a BottomFeeder suggestion the other day, and it seemed like a good (and simple) one to implement. I subscribe to a lot of feeds, but some of those feeds are more relevant to me than others. There are plenty of times that I'd like to see all the new items for the feeds that are most relevant to me - where most relevant means "ones I select most often". Well, I've implemented a simple tracker, and it's in the published (in Store) code. I haven't pushed it as an update yet, because I'd like to test it some more - I just made modifications a few minutes ago, in fact. It should show up shortly.
This is the second episode we recorded in Cincinnati at our internal "Camp Seaside" during the last week of September, 2007. In this talk, I got Alan Knight and Michel Bany together with Michael Lucas-Smith to talk about our upcoming support for Seaside in Cincom Smalltalk. Michel covered the history - how he came to the VW port of Seaside, and how far he's taken it. Michael discussed the automation of that process, and Alan covered what we're doing with Glorp for Seaside.
It was a good discussion, and we covered a lot of ground. If you have feedback, send it to email@example.com. Or, you can visit iTunes and leave a review, Podcast Alley and cast a vote, or find us in the Industry Misinterpretations group on Facebook. Make sure to watch for next week's episode - we spoke to Gemstone about their GLASS initiative.
After my recent trip to Cincinnati, I forgot to look over the log files. Last week's BottomFeeder downloads were proceeding nicely, at a clip of 243/day:
HTML tool distribution looks normal for this site, Mozilla heavy:
|Tool||Percentage of Accesses|
Finally, the Syndication numbers look normal as well:
|Tool||Percentage of Accesses|
|Google Feed Fetcher||6.6%|
|Net News Wire||6.5%|
|RSS 2 Email||1%|
In a political post on "Captains Quarters", I ran across a quote that applies equally well to product management decision making:
All elections are cost-benefit choices, at all levels.
The same holds true for product management decisions. In an ideal state of affairs, there's sufficient time and staff to fix all known problems with a product, and to build all the features that every customer and prospect you know would like to see. However, it's clear that we never have an ideal state of affairs. Time is finite - either a release ships within a reasonable timeframe, or its lateness makes people start to wonder (Vista, anyone?).
There's never really as much staff on hand as you might like - it's always the case that an extra person here or there could make a big difference. What that boils down to is the quote above, modified for product development:
All product development decisions are cost benefit choices
Since you never have the ideal, the best you can do is decide on what fits best with the time and staff you have: what will make the largest set of customers and/prospects the most happy. There's even something of an analogy to the way elections work: you can consider the primaries to be like your existing customer base: they want to see improvements in the existing system, while the general election is akin to prospects - you have to try and reach out beyond your base and appeal to new people.
It's not a perfect process in either case, and you always have people who are unhappy with the choices you make. The key thing to remember is that you can't escape that - any decision will upset some, and delight others. Sometimes you have no good choices, only least bad ones. Either way, you can't really escape by temporizing: failing to make a decision is itself a decision - in product terms, it usually means that you keep doing what you are doing right now.
Heck, that might even be the correct course - you often find out months or years later. Sometimes you never find out at all.
Technorati Tags: product marketing
On today's Smalltalk Daily, we load GLORP from the public store and take a look at a simple example with it. GLORP is an acronym for "Generic, Lightweight Object Relational Protocol". Which is kind of a mouthful - what it means is that GLORP is a framework for doing object to relational database mapping. Today, we take a (very) basic look at it.
Now that I've taken up cycling as an exercise, the last thing I need is morons like these to start making waves in my area. I just got involved in the local traffic advisory board, because I'm concerned that the "traffic calming" being done by the highway department is making it less safe to ride (never mind the fun it creates for emergency vehicles, school buses, and snow plows). The last thing I need is a bunch of self righteous idiots whose main contribution is to convince others that all cyclists are jerks.
I've published the "relevance view" that I posted on over the weekend, but only in the development update stream - if you don't see the update, you'll have to add "/dev" to the end of the update url in the settings file and restart the application. I'd be interested in any feedback on this.
I think I can say "Can't wait for Wednesday" over this:
No help for my lawn, either
I love the way Scott Karp oversimplifies and builds up a non-existant strawman over Facebook usage:
No, the issue is that so many “adults” fell for Facebook’s ploy to convince them that they should adopt a toy built for college kids as a platform for their professional networking objectives. The deep irony is that Facebook’s core student users -- for whom the application was and still is designed — are laughing at the grown-ups as they bumble around the playground, trying to hold meetings in the sandbox and forge new business relationships on the swings.
So are those student users going to drop off Facebook as soon as they get a job? Will they stop texting, too? Has it occurred to Scott Karp that most of us have more than one facet to our lives, and that - just perhaps - the one marked "business" doesn't have to be painted completely in shades of gray?
Mathew Ingram notes that some of the people in the music industry have a clue:
Fast-forward to today, and Rogers talks about how Amazon has finally created a music-download service that is actually as easy to use as a p2p network -- in fact, easier. Unfortunately, he says, it has taken eight years of wasted effort and millions of dollars in legal fees:
“8 years. How much opportunity have we lost in those 8 years? How much naivety and hubris did we have when we said, “if we build it they will come”? What did we spend? And what did we gain? We certainly didn’t gain mass user adoption or trust, two prerequisites to success on the Internet.”
I bet if you asked the RIAA, they would point to their recent p2p win and say that they've been vindicated. Which means that Rogers probably has a long wait before other people in music get it...
So much for that - maybe if Torre had started Mussina instead of Wang, things would have gone differently - but pitching is what cost the Yankees this series, which is fitting. I said earlier this year that they needed a better staf, and I wasn't wrong.
However, watch Steinbrenner get it wrong and blame A-Rod, and the rest of the bats that didn't wake up enough. Sure, A-Rod had a lousy post season - but if the pitching had been better, it wouldn't have mattered as much. Then again, if Torre had yanked Chamberlain as soon as it was obvious that he couldn't handle the bugs during game 2...
At the end of the day, the Indians wanted it more.
I think the RIAA p2p win is going to be looked back on as a pyrrhic victory in the end. The artists realize that the rhetoric and reality don't meet; the money all goes to the niddlemen. I expect to see more things like Radiohead's announcement - which is what Nine Inch Nails just did:
Hear that? It's the RIAA quaking in their diamond-coated boots as yet another A-list band gives labels the finger: Pretty hate machine Trent Reznor announced today that "as of right now Nine Inch Nails is a totally free agent, free of any recording contract with any label." Instead of futzing through the hapless middleman of an inept label, Trent's promising "a direct relationship with the audience as I see fit and appropriate," so we can expect more experiments in direct distribution and promotion, probably culminating in an album release not unlike Radiohead's In Rainbows.
The RIAA has been so intent on shutting down digital distribution that they haven't noticed what's been made clear: their disdain for their customers and their artists.
It seems that the version of Oracle I have installed is so old that some of the meta data GLORP relies on isn't in it. So, I'm installing a new database now :/
On today's Smalltalk Daily, we load the Active Record demo code (which is not completely baked - this is why the load sequence I'm using here will look a bit odd). GlorpActiveRecord is based on the object/relational mapping pattern ActiveRecord, popularized in Ruby on Rails. Today, we just get the demo set up - tomorrow we'll start doing stuff with it.
Well, there will be one less outpost of old-think in the PR game - Strumpette is retiring - perhaps tired of being shown up as an anonymous troll of astonishingly small amounts of intelligence. She won't be missed.
Nick Carr spots some interesting news: the online storage game is now offering guarantees in addition to nearly limitless cloud-space:
Nirvanax, an online storage utility now in public beta, has been touting its 99.9% uptime guarantee as a way to set itself apart from Amazon's S3 storage service, which has lacked a so-called service-level agreement. Today, Amazon responds by rolling out its own 99.9% guarantee.
He also notes that a new service, Flexiscale, is offering a 99.95% uptime guarantee. Storage is dirt cheap, but these kinds of service level agreements are going to offer value over the closets of cheap disk. Why try to manage your own power, cooling (etc., etc.) when you can pay someone else to do it - and get high reliability as part of the deal?
Technorati Tags: storage
On Monday, December 10th, there's going to be a "Smalltalk Day" Buenos Aireas, at the University of Buenos Aires. Two tracks are being held - one for educators and students, and another for industry applications. They are looking for presentations - you can get details (in Spanish) here.
Scoble notes that we really are in the narrowcasting business:
Every time I get on TechMeme I get 500 to 3,000 visits. That matches what the Guardian and what Nick Carr are seeing.
But, truth is not many sites out there do any better. Yeah, when I get on Digg I get 20,000. When I got on the front page of the BBC a couple times in the past month I got 5,000 each time. But Valleywag? I get 100 to 1,000 visits (I’ve been on there something like 20 times including with some VERY sensational posts that would make anyone click and ask themselves “what the heck did that guy do?”)
I certainly don't see a noticeable pickup from being part of the Techmeme discussion links - and this site lands there a lot. Unless you're writing a celebrity or political blog - and even then, you have to get noticed first - you're getting mostly a set of "regulars" to your site. There's nothing wrong with that, but it's worth remembering the next time you look over your logs.
I might have been on to something with my thought that the recent RIAA p2p victory would be seen as pyrrhic. Techdirt notes that a small, but growing group of artists are starting to figure out that the "value add" of the music labels isn't such a great deal:
There are two key things to note in all of this. First, all these bands feel the need to ditch big record labels to do this (and, no, that doesn't mean that small bands without recording contracts can't succeed this way too). This is a sad state of affairs for the record labels -- because there still should be a place for them in helping to promote and market a band, even if they're giving away the music for free. It's just that they're not venture capitalists any more and bands don't need help in distributing content -- two businesses the record labels insist they're in.
I think this only fuels the desperation and stupidity on the part of the labels. You've got a whole ton of useless middle-men who are accustomed to a comfortable existence - and digital distribution has come in like a sledgehammer. They no longer control the horizontal, the vertical, or anything at all. Bands can promote their music in a variety of online forums, and not sell their souls for a few pennies of royalties.
Sure - most bands will have to rely on touring to make most of their money, but that's no different than it is now. The real difference is that the incremental revenue they do get from online sales of mp3 and CDs will go in their pockets instead of someone else's.
Technorati Tags: RIAA
I found this in a comment to an unrelated post, so I figured I'd highlight it and address the issues. He also complained about the installer; I sent that to the relevant engineer here at Cincom:
My real frustration comes from trying to install packages from the sTore. I wanted to install Pier, so I went to the pier website, read the installation instructions, and it says to load the pier package from the store. So I did that, yet the installation crashed because magritte was not loaded. So I tried to load magritte, yet the installation crashed because seaside was not loaded. So I tried to load Seaside, but the installation crashed because of some kind of 'dllcc' problem.
Perhaps I should continue looking into this issue, maybe I made a mistake somewhere. Yet, when installing packages, I'm used to it either telling me what I need, in plain english so that I can get the package, or giving me the choice of installing it automatically.
This has to do with pre-requisite settings (or, in this case, the lack of them). I tried loading Pier myself a few weeks back; like the commenter, I gave up in frustration. Now in the case of Pier, I know that the VW port is a work in progress, but - at the same time - having load problems makes it very difficult for others to pitch in. This is a problem with plenty of things in the public repository (I've released code myself that is hard to load). The bottom line is: we (Smalltalkers) need to remember that there are new people coming in all the time, and having something break completely on load is very offputting.
Answers.com is now using a widget for travel that takes you to a Seaside application - do a search like this one, and the reservations application after the widget is Seaside. (Via Squeak News). (Corrected based on the comment that came in).
Incidentally, I did a podcast with Liz Cohen of Answers.com awhile back.