Smalltalk Tidbits, Industry Rants

Here's an all too typical story in SD Times - writer Jennifer deJong lets a Forrester analyst say something utterly absurd:

“In the early days of agile, only heretics used tools,” said Liz Barnett, a Forrester analyst who follows agile development. But that is no longer true.

Umm - XP, which is where the entire agile movement comes from, was developed by Kent Beck and the C3 team. Hmm - what were they using on that project? Smalltalk (VisualWorks, specifically). Last time I looked, VisualWorks is a tool filled environment.

How frelling hard would it be for a journalist or analyst to do some minimal research? Is typing into the Google box too hard for these people?

I hope that Sci Fi does better with this series than they and Fox did with Sliders, which started with such promise. I love alternate history books (like the ones Turtledove writes). Here's the blurb:

SCI FI Channel announced that it is developing What If, a "speculative future" series that poses intriguing scenarios of alternate realities. The program, from NBC Universal Television Studio in association with New Line Television, asks, "What if a moment in time could change the world forever?"

I understand what Stallman is after with his comments here. I dislike the way DRM and software patents are going as well - and have said so here numerous times. However, what's the likely result of changing the GPL to penalize patented software and DRM? Expect a lot more firms to develop strong allergies to the GPL

I blogged about this Javacast the other day, and I finally got around to listening to it today. Tune into it at minute 1:22. The guys describe a small web app that they built using a "lite" Java stack - Spring and Hibernate instead of the full J2EE stack. Time to build - 4 months. Ok, they decided, on a whim, to try Ruby on Rails. Time to build?

4 Days

That's a pretty amazing productivity jump, and they were pretty shocked. They went on to state that dynamic languages just have it all over Java in this sphere, mentioning Ruby, Lisp, Python, and Smalltalk (stating that Ruby has momentum).

Then there's an interesting riff starting at minute 1:26, where they go into Continuation based web development, and they point out how much simpler things like Seaside are - they went so far as to say that they figure that in 3-5 years, we'll all be doing development like this. They mentioned Seaside and Squeak specifically (note - Seaside works in VW as well). They point out that you can't get there in Java, because you can't do Continuations. Instead, there are frameworks that build huge state machines (complexity alert!).

So, if you want the power and productivity now, you can grab VW or Squeak and start using Seaside. Or you can get the all too typical pale reflection in Java...

There are more influences out of Smalltalk than most people realize - take the Eclipse Foundation, for instance - 1/3rd of the fulltime staff came out of Smalltalk.

Hat tip James Governor

Of course, those of you want the real deal instead of the pale reflection can download it :)

We sure don't need this: that system of disturbed weather off the Florida coast has turned into Tropical Depression 16. Follow the link to see where it's headed. We need this like we need a hole in the head...

Over the last few weeks, I've noticed that Eudora was getting very sluggish, and would often tie itself up for minutes at a time. Seems I was keeping too much mail around for it to handle in any kind of reasonable way. I subscribe to a bunch of mailing lists, and get the all too typical spam load - so I receive well north of 300 emails per day. I hadn't deleted any mail since 2003, and Eudora was having trouble with folders that had tens of thousands of messages. So... I whacked all the older ones (after saving copies), and Eudora was back to its formerly snappy self.

Anyone know if Thunderbird is able to hold large amounts of mail without this issue?

I like Graham's latest essay. My readers know that I don't often think highly of Graham's essays, but this one caught my eye. Agree or disagree, it's thought provoking.

The National Hurricane Center is tracking Tropical Depression 15. It's east of Florida, and could track across the state into the gulf. Of course, it might just break up and go away too, there's no telling yet. Here's hoping it breaks up - the gulf coast doesn't need another event right now...

So my wife wanted to export her mail - the messages and account settings - from Outlook Express on machine 1 to Outlook on machine 2. You would think this would be easy - same network, all MS apps - but no, it's a two step dance of pain. Here's what we ended up doing:

• Zipping up the Outlook Express directory structure (under My Documents)
• Sending that to Machine 2
• Trying to import the result into Outlook - nah, that didn't work. Outlook didn't claim to recognize it
• Google for help, find MS support pages that are helpfully useless in anything but IE
• Finally realize that the step to take is to import into Outlook Express on machine 2, at which point Outlook can import from Outlook Express.

Sheesh. It was fairly quick once we figured it out, but there was no good reason for us to have to take the side trip through OE on machine 2. Someone needs to find Ballmer and crew and point out the term "Usability". It's pretty damn clear to me that they haven't heard of it.

Heh - even better. We finally got it all done, and my wife actually looked at Outlook. Her reaction makes me look downright positive about that abortion of an application. The user interface is more cluttered than I remember it - like many MS apps, it seems to have gotten worse with age. So now it's off to find a mail client that isn't cluttered as all get out and can import from Outlook...

I've found a few more issues with the new tab support - changing Look and Feel caused startup problems, and you could get errors when clicking near (but not on) a tab. There were a few other things as well - an update is available in the dev level updates. bear in mind, if you don't have the latest dev updates (with tabbed browsing), this is all something you can ignore for now.

I also fixed a browser launch bug on Unix/Linux. In some shells, urls with ampersands were being taken badly - the ampersand being seen as a request to background the task rather than as part of the url. I fixed that, and the person I tested it with reported that it worked just fine.

I love seeing stuff like this:

Seaside is incredible too, I picked it up very quickly reading 3 pages of overview, playing with demos, and then on to coding. There's plenty more to learn, I'm sure, but my web app is up and running with suprisingly little code. Learning/using Java frameworks was never this easy.

The environment in VisualWorks is awesome - integrated refactoring browser, code critic, sunit testing framework - it feels like home, one I never want to leave again. Unless something more like Smalltalk than Smalltalk comes along. :)

Wilfred Springer clearly hasn't seen the refactoring tools in Smalltalk, or he wouldn't have let loose with this:

But I also remembered some conference sessions on dynamicly typed languages that I attended in the past. None of these sessions ever mentioned the fact that refactoring tools can be so much more powerful if the type is known before runtime. Renaming an operation in a large codebase would be practically impossible. Tracing the usage of a class would be equally hard.

Seems the Java guys are starting to realize that there are simpler ways to solve problems - this weeks JavaCast talks about Seaside, amongst other things. Very cool.

Dave Winer noticed something interesting about Technorati's new blog search - various tags that you would think would be associated with a given blog aren't. This is related to something Tim Bray commented on the other day - the current rankings are reflecting who has and hasn't charged in and claimed the 20 tags that Technorati lets you associate with your blog. Over time, that should sort itself out - what we are seeing right now is the equivalent of a sorting algorithm after the first few iterations.

If you've grabbed the development build or updates, and have tabs - there are some areas where you can run into problems. I'm running the dev build myself, so I'm fixing issues as I run across them. Keep in mind that the development builds are not necessarily completely baked.

Browsing around, I discovered that the city of New Orleans has a disaster plan - i.e., they had a stated plan of action for this kind of situation. One of the sadder things I'm seeing right now is a desperate attempt by partisans - of all stripes - to grab this disaster and use it as a stick with which to beat their opponents. I'd guess that mistakes were made up and down the chain here:

• Bureaucratic ones further up (Washington), because the larger and organization gets, the more paper is required for anything.
• Local failures to follow the plan - because a plan is one thing, an actual emergency is another.

On that latter point - take a trivial example, backup of crucial data. We all know that hard drives fail, and that we'll face a loss of data at some point. We know this. And yet, how many of us (myself included) fail to do backups diligently, subconsciously thinking "it won't happen to me"? Now, when we do lose data, how many of us decide to point the finger at IT, since it's far easier to do that than look at ourselves?

On the first point, how many of us know a "stickler" who insists on following "the rules", no matter what the situation is? Bureaucracies are loaded with that type of personality.

There's a lot of finger pointing happening here, with the addition of partisan game playing by all and sundry.

I have a few more shots from today's outing - here's one my wife took at sunset:

Now, here's a nice shot my wife took as we were out on the water. You can click through this one for a bigger version:

We had a treat this afternoon and early evening - my cousin Eric works with a boat owner, Dave - here's a picture of the boat that my daughter took. This pic is from after the outing we took with Eric, his son Chad, and Dave:

So, back to the beginning - we met Dave a month or so ago, before we went on our trips - Eric told us that he was interested in learning Puerto Rico. We didn't get to that game that evening, but we got a very pleasant boat outing this evening - and we'll get to the game at some point in the near future.

It's a very nice boat, and it was a wonderful day for an outing - sunny, not much breeze, clear blue skies. Here's an idea of what the day was like - my daughter took this while we were out:

We sat up on the prow of the boat for a fair bit of the trip - my daughter and nephew loved it up there, so Eric and I got out there as well. You get a great view of things from there, and a good breeze, depending on what the weather is like. Here's a shot of Eric and I up there:

After we pulled back into the pier, we sat upstairs for a bit and chatted before dinner. While we did that, Victoria got this shot of the sunset over Baltimore and the harbor:

I'll post a few more shots in the next post - my wife is just pulling some pics off of her camera.

This is so true that it hurts. Read this, then ponder the next few ads you see.

New Orleans will be rebuilt, but it will never be what it was. Consider - the process of:

• Plugging the Levee holes
• Draining the water
• Inspecting/Demolishing buildings

Will take many, many months - I'd say longer than anyone is guessing right now. The historic district seems to have mostly escaped, and will be one of the first places to come back - property owners there will actually have property to come back to. The rest of the city?

Well, consider - there's some proportion of the population (many of the 80% who evacuated, I'd guess) who have family and/or friends who will be able to take them in. For those people, regardless of their age, this will be like a return to being 18 (or 21, if they went to college). They'll have to start all over again, with nothing but a helpful push from their families. How many will return to New Orleans? Not many, I'd guess - most won't want to put their lives on hold for the next 6-12 months.

Then there are the people who are only being evacuated now - they have even less than the first group. Many of them don't own homes (I saw census bureau data that stated a 49% rental rate in the city). They've lost almost all the goods they did own, so there's very little holding them to the city - no jobs, no family - nothing. Unlike the group above, these folks probably don't have as many outside family/friends to go to (if they did, I suspect they would have gone there earlier). This means that they'll start without the helping push, but they'll have to start again nevertheless. And like the first group, I very much doubt that they'll sit in a holding center for 6-12 months when there's nothing to return to.

So what do have left? The residents of the historic district (or buyers of their property in the interim - there will be speculative buying on the cheap going on). Business interests that need to be there - import/export, energy, fishing. However, that's going to be limited by the willingness of insurers to sponsor construction - and believe you me, those outfits are going to take a very critical eye toward rebuilding in areas that are under a lot of water. Regardless of what new plans come down the pike to rebuild levees stronger, everyone knows that it will be a multi-year effort to do that - and any new construction will be in danger while that happens.

Consider the history of Galveston, TX after the ruinous 1900 hurricane:

While Galveston received financial help from the county, state and federal governments, a large portion of the burden had to be carried by the city itself, at the expense of other projects.

McComb sums it up about as well as it can be:

"Human technology made it possible - for the city of Galveston to remain on such unstable land. The city did not flourish. Houston - left the island city far behind. Galveston simply survived.

That's the future history of New Orleans, right there - and bear in mind, that as bad off as Galveston was then, New Orleans has it worse. Galveston, like Biloxi today, merely (I hate to put it that way) has rubble to clear before rebuilding could start. New Orleans has all that water...

It's time for my weekly look at the logs. First up - the BottomFeeder download report:

Platform	BottomFeeder Downloads
Windows	1877
Mac 8/9	691
HPUX	414
Linux x86	355
Mac X	305
Sources	256
Update	124
CE ARM	117
Windows98/ME	104
Solaris	21
Linux PPC	11
SGI	11
Linux Sparc	6
AIX	5
Source Script	5
CE x86	1

Those numbers are still up at the 600+ per day level - while I was traveling, I must have gotten noticed somewhere. That's cool, since the next release is imminent. Next, the HTML blog accesses - see what the tool distribution looks like:

Tool	Percentage of Accesses
Mozilla	44.5%
Internet Explorer	35.1%
Other	10.6%
MSN Bot	6.8%
Google Bot	3%

Either my readership is getting more diverse (and the raw numbers are up some), or MS is making headway with IE - or, looking below, the MSN Bot is still creating churn. Finally, the distribution of tools on the RSS feeds:

Tool	Percentage of Accesses
MSN Bot	24%
Mozilla	17.9%
BottomFeeder	11.6%
Net News Wire	9.3%
Other	7.8%
Safari RSS	4.1%
BlogSearch	2.9%
Internet Explorer	2.6%
BlogLines	2.4%
NewsGator	2.2%
SharpReader	2.2%
Planet Smalltalk	1.8%
Feed Demon	1.6%
Magpie	1.5%
Lilina	1.1%
RSS Bandit	1%
Liferea	1%
Feed Reader	1%
Feed Tagger	1%
JetBrains	1%
RSSReader	1%
News Fire	1%

Looks like the MSN Bot is still creating lots of churn - there is no way that it should be responsible for that much of the RSS grabbing - not two weeks in a row. The gnomes in Redmond need to tweak that puppy...

I didn't think that a columnist could surprise me with raw ignorance any longer, but I was wrong. Have a look at the lack of understanding on display here - eWeek's Jim Rapoza is utterly baffled by syndication and aggregators:

But after years of really trying to like RSS feeds, I'm finally waving the white flag. This is another technology that just isn't for me. (And, yes, I realize that some of you are reading this column after receiving it through an RSS feed.)

As I do with IM and cell phones, I think part of the problem is that feeds— while empowering in some ways -- also remove some of the user's independence. Once subscribed to a feed, you just keep getting it, whether you want that day's info or not. No matter how I get the feeds delivered, they eventually become noise on my desktop or even an actual nuisance that I'd rather not deal with. Feeds delivered through e-mail clients are the worst, as they eventually become associated with other e-mail nuisances, such as spam.

My adventures in RSS feeds have succeeded in making me really appreciate Web browsing. I've decided I much prefer to go to Web sites and blogs to see if there's new information I want to read that day. I just feel as if I have a lot more control that way, and, using the grouped bookmarks features in Mozilla and Firefox, I can quickly look at multiple similar sites.

It's actually hard to stumble across this level of ignorance without trying. It's one thing to dislike the potential overload from RSS - I tend to max out around 300 feeds, myself. But Rapoza just doesn't understand the basics. You get the data whether you want it or not? Which part of "unsubscribe" is too hard for this guy? In my aggregator, I select a feed, and then pick the "remove feed" menu option - bingo, no more data from that source.

Apparently, Rapoza thinks that Winer and Scoble are going to show up at his house and throttle him if he dares to unsubscribe. And they pay this guy to produce drivel like this...

Now that I've got the major features for BottomFeeder 4.0 packed away, I've got another task coming up on my plate - comment registration for the Silt server. I'd like to run this as follows:

• Continue to allow anonymous comments, but such comments will go through the fuill spam check gauntlet
• Add Registration, where anyone who registers and then logs in can comment without the full spam check

The main question is this - should registration be site wide, or blog specific? I'm still pondering that one.

GNU Smalltalk 2.1.12 has just shipped, and Instantiations has delivered VA Smalltalk 7. Looks to me like there's a Smalltalk solution out there for every need and desire :)

Blaine has the lowdown on the upcoming Dolphin 6 release. Looks like Andy and Blair have been busy :)

Bob points to something I missed today - the Adventa success story. Here's what Adventa says about Smalltalk:

Yes, there are more engineers in the world who know "C", there are more who claim to know C++ and "Visual this" and "Visual that" but the fact remains, and we prove it every day, there is not a software engineer anywhere, working in any language, that is more productive than a "Smalltalker." It was obvious to us in 1988, in the year 2002, there is still "none-better." JAVA? Maybe, someday. Right now, JAVA is many years behind Smalltalk in terms of maturity, capability, reliability, etc. It's good for lightweight apps, but can't match Smalltalk for productivity and reliability.

There's a dev update for BottomFeeder posted with tabbed browsing in. I'm about to start a build, so that I can push up new builds that have the latest stuff, without the need to grab updates. There's a new tab item on the toolbar, as well as a menu pick. Each tab also has a small menu allowing you to tear a tab off or close a specific tab.

I would have had this done sooner, but it took me a bit to realize that menus for tab items were an extension added by TypeLess (the IRC plugin) - and I was missing a crucial (but not so obvious) bit of the API. I got that handled, and it's working nicely now.

Update: The downloads are available under the development links now

Dare Obasanjo sounds pretty disenchanted with working at Microsoft:

A good example of this is taking a look at Windows from the consumer perspective. The decisions that Microsoft has made over the past couple of years from abandoning feature work in Internet Explorer until Firefox became popular to a lot of the original intentions around the 3 pillars of Longhorn (Avalon, WinFS & Indigo) are the actions of a company that is more interested in protecting its market share than one that is trying to improve the lives of its customers by building great software.  Of course, it's not only customers that get the short end of the stick. Employees also have the consequences of this kind of thinking to deal with as well. The primary way this manifests itself is integrated innovation, a buzzword that translates to more dependencies among shipping products, less control of one's product destiny and longer ship cycles. A lot of the frustration you see in the comments in places like the Mini-Microsoft blog are a direct consequence of this focus by our executive leadership.

The amazing thing is, he intends to stay there for the short term. The above is an amazing airing of "dirty laundry" - and, I think, a bridge burning exercise.

I've got a brief look at ObjectStudio in VW here - this is very early access stuff - I got walkbacks doing some things. In any event, a couple of screen shots to show you that things are progressing:

That's the menu pick showing ObjectStudio inside VW. Now, the next one is OST after it's been fired up:

Just a brief glimpse, and there's a long way to go on this - but it's progress

I'm pretty close to the 4.0 release of BottomFeeder. I've got tabbed browsing in, but there's some wonkiness in the menus for the tabs that needs to be worked out. It's something simple, I'm sure - but I've got bunches of other things distracting me today. Amongst other things, I'd like to post some screen shots from the ObjectStudio in VW work I referenced the other day - but corporate is having some VPN connectivity issues right now, and I can't get to the internal download sites.

Additionally, I'm working on various and sundry other small projects. I should have the problem sorted out soon, and then it's just a matter of documentation - there are obvious UI changes associated with the new tabbing.

I ran across an interesting analysis of the flood issues in New Orleans in "Civil Engineering" - written well before the current disaster. There's a ton of information in this about the danger that faced the city, and what the answers had been over the years. The bottom line on this is, the plan for anything as powerful as Katrina was "pray it doesn't happen". Consider:

In 1999 the Corps was authorized by Congress to study the feasibility of various proposals for protecting the city against such devastating storms. An obvious possibility would be to raise the current levees to a height deemed acceptable by an AdCirc analysis. That, however, would also require widening the levees, which may not be possible in many areas because of the proximity of homes. Among other alternatives, Naomi will investigate the possibility of creating an immense wall between Lake Pontchartrain and the gulf to keep water out of the lake during a severe storm. Such a project would involve constructing massive floodgates at the Rigolets and Chef Menteur passes, where storm surge would enter the lake.

According to Naomi, any concerted effort to protect the city from a storm of category 4 or 5 will probably take 30 years to complete. And the feasibility study alone for such an effort will cost as much as $8 million. Even though Congress has authorized the feasibility study, funding has not yet been appropriated. When funds are made available, the study will take about six years to complete. “That’s a lot of time to get the study before Congress,” Naomi admits. “Hopefully we won’t have a major storm before then.”

Forget levees for a moment, and consider the other problem that is busily creeping toward New Orleans - the loss of coastal marshland. The taming of the Mississippi has meant no new floods as bad as the 1927 horror, but it's had a nasty side effect - the protective buffer of marshland (the bayou) south of the city is disappearing, fast. At current rates, New Orleans will be a gulf coast city within 30 years. That would mean that - notwithstanding the current flooding - Lake Ponchartrain would be the least of the problems facing the city - a similar storm hitting in 30 years would visit the storm surge that wiped out Biloxi and Gulf Port on downtown New Orleans:

“We’re trying to enforce human decisions on a natural process,” says Naomi. “What we’re trying to do is take a snapshot of geologic time and say, ‘This is what we want; this is where we want to live.’ The question is, Is it going to be feasible in the long term?”

Naomi says this question will not be answered with levee feasibility studies alone. It will also require a more complete understanding of the natural processes at work in and around New Orleans. For example, the wetlands of coastal Louisiana, which would act as a buffer and slow any storm during its approach to the city, are dying because the freshwater and nutrients that historically flooded into them from the Mississippi can no longer escape the river. At the same time, the sediment deposited here by the river long ago is subsiding, and no new sediment is overflowing to replenish it.

That's not the worst of it, believe it or not. The entire southern portion of Louisiana is sinking:

The true situation, however, is almost too grave to consider. “Coastal Louisiana is sinking under its own weight,” Dokka says. “The ground in Louisiana is ultimately going to go under.”

Indeed, the state is subsiding so quickly that the National Oceanic and Atmospheric Administration’s National Geodetic Survey (NGS) considers the orthometric markers in Louisiana surveyed every decade for the North American Vertical Datum (NAVD) to be “obsolete.” Dokka and his colleagues, together with experts from the NGS, are now using high-powered transponders and numerous Global Positioning System satellites to develop “true” elevation points in the state on the basis of their relation to the center of the earth.

Using a rate of subsidence measured at a tidal gauge off the coast of Grand Isle adjacent to an original NAVD marker, Dokka was able to calibrate rates of subsidence at hundreds of other markers around the state. His results indicate that the elevations of some areas have dropped as much as 2 ft (0.6 m) since they were last surveyed for the NAVD. Based on Dokka’s “true” elevations, some of the Corps’s levees in New Orleans may be more than 1 ft lower than their posted elevation.

Which means that the entire levee system is just so much whistling past the graveyard.

New Orleans has been in an exposed condition for well over a hundred years. This time, the city's luck ran out. The main thing we can do now? Visit this FEMA page, which lists charitable organizations that are helping out. Donate to or volunteer with one of them.

Jon Udell is posting on the RSS redirection issue:

It's been almost three years since I first wrote about the problem of RSS feed redirection. From time to time I'm reminded that it's still a problem, and today I noticed that two of the blogs I read were affected by it. I was subscribed to John Ludwig at www.theludwigs.com/index.rdf, and today's entry says "Feed moved -- pls check out www.theludwigs.com/index.xml." In fact he's got an index.xml and an atom.xml, and the latter seems to correspond to what's actually published on the blog, but either way the issue is that we've still yet to agree on a standard way for newsreaders to follow relocated feeds.

Well, there's a solution already - HTTP redirect. I know, I know - as Jon says, not everyone controls their own server, and not all hosting providers will provide that service. Still, it seems like the KISS solution would be to advocate for hosting providers to agree to N days (weeks, whatever) of standard redirect service followed by a 410. Solves the entire problem, most aggregators already handle redirects - and it doesn't require the invention of funky one-off, feed level answers.

What am I missing?

Here's some live-blogging from New Orleans - far better feel for what's actually happening on the ground than the media reports I'm seeing.

And in the "who would have thunk it" category, here's a network vendor still on the air from the city - up in a tower with supplies and generator power.

Derek asks the question that's been in the back of my mind, and I'm sure it's been in the back of other people's minds as well:

So now that New Orleans is underwater, and the devastation of property is huge, various charities have started beating the drums about raising money to "rebuild New Orleans".

I'm forced to ask myself one simple question: "Why?"

As he says, it's not as if this threat was unknown - it's been the "crazy aunt in the attic" of New Orleans for decades. It's worse than the coastal communities that get whacked from time to time, because at least the water leaves those - as bad as Biloxi was hit, for instance, it's not underwater.

CNet has a story about yawning - the whys, wherefores, etc. Check it out, and see how many times you yawn while you're doing so :)

Tim Bray makes some assumptions about threading and scalability here:

In that recent Ruby piece, I remarked that Ruby threading struck me as kind of feeble, and that threading is getting real important. Well, I know one way to solve that problem. So I tracked down JRuby geek Tom Enebo and got some news and he pointed to me to some code that I think is pretty cool.

On that threading thing, Tom tells me that “We currently implement Ruby threads using Java native threads. It is our plan to continue doing so.” So, JRuby is ready for the threading era.

That may not be the best way to handle it. We actually have some good experience in this area, due to the port of Opentalk (our communications framework) to ObjectStudio for the last release. ObjectStudio maps its threads down to platform (in this case, Windows) threads. That makes them non-deterministic and much harder to control. It also means that any error made by a developer in an individual thread has the ability to tie things up very nicely - at the platform level.

In VisualWorks, threads are green. This makes them really, really inexpensive. For instance, in BottomFeeder, I spawn a thread per feed (I'm subscribing to 309 as I write this). I don't need to worry about thread pools, or overwhelming the platform. If I tried such a feat with native threads, I'd have to worry about those things.

Sure, you'll say, but what if I have a multi-CPU (or multi-Core) box that I want to take advantage of? Simple - run more than one image (process) and have them communicate as needed. That's actually the classic Unix approach to scaling, and it works quite well. The programming model is vastly simpler - I don't have as many complex synchronization issues to deal with. External APIs (such as database calls, etc) can already be threaded at the platform level, so you don't have a blocking issue.

Update: Patrick Logan has some thoughts on the issue.

I just added some simple (and obvious) modifications to the posting tool that ships with BottomFeeder - obvious keyboard shortcuts:

• shift-insert -- paste
• shift-delete -- cut
• ctrl-b -- bold on/off
• ctrl-i -- italic on/off
• ctrl-u -- underline on/off
• ctrl-l -- hyperlink on/off

All things I should have added a long time ago, but hey - better late than never :)

Holy smokes - I was flipping between Fox and MSNBC this evening, and both networks had interviews with a spokesman from the Army Corps of Engineers. It doesn't sound good. While the water level in New Orleans is no longer rising, that's only because the water has reached its natural level. Apparently, they just found more breaches in the flood walls that they weren't aware of, and they've not been successful in their attempts to block them. Why? They aren't accessible by land or water, and they can only do so much from the air.

It gets worse. Even once they get the holes plugged, they then have to drain the city. The spokesman said that would be a 3-6 month job. So what we have is a nearly unprecedented situation in the US - hundreds of thousands of refugees who won't be able to return home for months. Many of them may not ever get home. Consider the buildings in flooded New Orleans. First, there's whatever wind damage they took during the storm. Add in a few months of being marinated in semi-toxic water. What do you end up with? A whole lot of buildings that need to be condemned and demolished, that's what.

Things are going to stay bad for New Orleans for a long, long time.

This is a Q/A session with the Army Corps of Engineers - and it sounds like the "Best Case" scenario is pretty tough. Never mind that the caveats thrown out are pretty big. There is simply no telling when - or, to be really pessimistic if - New Orleans will recover.

It could get worse too - have a look at this story - here's the kicker:

Once the levees are fixed, Maj. Gen. Don Riley of the Army Corps of Engineers said, it could take close to a month to get the water out of the city. If the water rises a few feet higher, it could also wipe out the water system for the whole city, said New Orleans' homeland security chief, Terry Ebbert.

Thus far, they have had little luck plugging the holes in the levees - and the water is still rising. I understand why the Governor of Louisiana has a "deer in the headlights" look when she speaks in public - it must be horrible to be in charge of a state, and be utterly powerless against this sort of thing. All we can do now is donate to the Red Cross, and pray that the experts can get those holes patched.

Patrick Logan is not impressed with the WinFS effort, saying that the WinFS team is stuck in outdated thinking.

Susie is based on Little Smalltalk, updated and aimed at specifically at scripting. Have a look

Doc has lots of details on the state of the Gulf coast - and it's not good. Something like 80% of New Orleans is underwater, and it sounds like the attempt to use giant sandbags to patch the biggest levee breach have failed. It's going to be a long, slow recovery for the gulf coast - and not all of it will come back as it used to be.

We have an interesting project going on right now - we call it ObjectStudio in VisualWorks. As you probably know, Cincom Smalltalk consists of two Smalltalk environments - ObjectStudio and VisualWorks. Most of our customers use one or the other, but we do have a number now who are using both. That number is going to increase soon, I expect.

ObjectStudio is a pure Windows development environment, and is usually used for Client/Server projects. VisualWorks is cross platform, and suitable for a wider range of projects. The biggest difference between the two has been at the VM level - VW has a modern, JITing VM which is very fast, and very scalable. ObjectStudio, on the other hand, has a slower, purely interpreted VM. Additionally, an awful lot of things that are implemented at the image level in VW are implemented at the VM level in ObjectStudio.

Which explains where we are taking ObjectStudio - into VisualWorks. ObjectStudio will ship as a separate image (we also plan to make it loadable), hosted inside the VisualWorks environment. That's going to entail some changes for ObjectStudio developers, but it will also entail a number of new capabilities - every feature of VisualWorks will be available in ObjectStudio, and every feature of ObjectStudio will be available in VisualWorks.

There will be some changes - at present, ObjectStudio threads map to native Windows threads. In the OST in VW release, ObjectStudio threads, like VW ones, will map to Smalltalk level green threads. Using the VisualWorks VM, ObjectStudio applications will be a lot faster. There will also be some code changes necessary - we will have a white paper detailing that list ready shortly.

The good news is this - the key features of ObjectStudio:

• Native Windows GUI
• Database Modeling Tools
• OLE embedding

will all be available - just as importantly, these will all be available for VisualWorks developers as well. This is not simply a change and upgrade for ObjectStudio developers, but a set of new capabilities for VisualWorks developers as well.

Stay tuned for the white paper - I'll have links once it's ready. As to timing? We may have early access (meaning, pre-beta level) available this winter. As engineering moves through the development cycle, I'll have more details.

ESUG has the answers - check it out

I've posted this before, but the delicious Smalltalk page I set up is a great place to find Cincom Smalltalk related links.

A lot of the early reporting on Katrina sounded like this - "it was bad, but whew! New Orleans dodged a bullet".

Well, it looks like that sentiment was expressed too soon. Apparently, at least three levees have failed, and New Orleans is filling up like a soup bowl. Take a look:

Last night, MSNBC had reporters outside the Superdome, looking dry. It looks like my earlier worry about getting people out of the Superdome might become a live problem. From the MSNBC reporting:

Some hospital patients were airlifted to the Superdome as a precaution, and NBC's Brian Williams reported from the edge of the French Quarter that what had been dry land “was filling with water” by 5 a.m. ET. “We have a new problem in this city,” he said. By 8 a.m., some French Quarter streets were under several inches of water.

NBC's Kerry Sanders, reporting from a helicopter above the city, said “it's basically one giant lake here in New Orleans.”

It's going to be a long time before they get this straightened out. To see the problem, go watch this Nova video, which was produced earlier this year, in the wake of the Ivan near miss.

Update: Here's a news feed from one of the New Orleans area TV stations. Lots of updates, both text and video.

If this NY Times story is right, we have the killer appliance in the wings - a cell phone/iPod combo. Add in a camera, and boom - it's a one-stop category killer, IMHO.

Ted Leung is right on the technical issues here, but not on the wider buzz:

Lots of people are up in arms about IM clients. People are complaining about the Windows client for Google Talk. People are complaining about the lack of Mac and Linux clients. But I think that this is misguided. The real value and potential in Google Talk is that Google might be building an open, un-siloed IM network. It will be interesting to see whether "don't be evil" actually works in this case, because starting another closed, siloed IM network would definitely count as evil (not to mention pointless) in my book.

To be specific, the bleeding edge technology-philes are "up in arms". The users don't care. I say this based on one of the main target customers for IM technology - middle school and up age kids. My daughter is an avid IM user, as are most of her friends (the exceptions being kids with parents who don't allow it). The lot of them use AIM. I've never heard them complain about not being able to chat with so and so due to the silo problem - all the kids just download AIM and move along.

Which is why there's no real pressure to fix this - from the standpoint of the vendors, there is no problem - our rantings on the subject just don't matter to them.

Some companies are filled with slow learners. Witness Dell - they are still adjusting to the "Dell Hell" storm from Jarvis and company - and they are dragging their feet on a maintenance issue:

In most businesses, when a product is discovered to have a defect that's likely to cause it to fail, the manufacturer issues a recall for the affected units. Is there any reason that shouldn't be true in the computer industry as well? That's what many IT mangers with large installations of Dell Optiplex GX270 systems have to be wondering right now.

How many times do they need to get slapped before they realize that a blog storm isn't good for business? At least in this case, it seems like the answer is: a lot. If you see your company heading down this road, be very afraid...