Smalltalk Tidbits, Industry Rants

This list is too true :) Via Dana VanDen Heuvel.

Doc Searls asks a question about the Times Select policy (where all the Times columnists are now behind a pay wall) - and then tosses out a potential answer:

Question: Why give away online what you charge for in the daily paper, while charging for what becomes fishwrap tomorrow? The Times' answer is to charge for both op-ed writers (why op-ed and not the rest of the paper) and fishwrap. This whole thing smacks of a political compromise between warring factions inside the paper.

I had two other possibilities come to mind as well:

• A management clue breakdown - "We're the New York Times - of course everyone will pay!"
• The paper's financials are far worse than anyone outside thinks

I suspect we'll find out soon, because the market for political opinions commoditized a long time ago...

One of the things you bpick up around the blogosphere is the huge wads of self importance. I'm as guilty of it as anyone else - I've given whole talks on how crucial it is to pay attention to blogs. Having said that, there's another side of this, and it impacts tech bloggers and political bloggers alike - the "echo chamber" problem.

It's very easy, watching Technorati, and Memeorandum, and IceRocket, and Google's blog search, to get caught up in any of a number of ongoing conversations. Take a look at Scoble's latest comparison post, for instance. You see this kind of tail chasing all across the blogosphere, although it seems positively endemic on the political sites - I guess the egos are just that much bigger over there.

There's an easy reality check though, and I get it every time I visit customers. I ask "How many of you read my blog?" Now, as egotistical as that sounds, stay with me a second - in these meetings, I'm seeing Cincom Smalltalk customers, and a goodly proportion of what I (and the other bloggers here) cover is Cincom Smalltalk focused - so the cst blogs are pretty "on topic" for a customer crowd. The thing is, it's always a fraction of the audience that is reading.

Most people just aren't as focused on their jobs (or politics) as bloggers are. We tend to be obsessive about the topics we're interested in, posting multiple times a day. Most people don't spend even a fraction of that much time pondering the industry they work in or day to day politics. They have other things on their minds - family, sports, hobbies - other things that fill their time and make them happy. Meanwhile, bloggers - in any part of the sphere - will form a circle of self importance and blather about the huge impact they have.

Don't get me wrong - within an influential sector of the population, bloggers do have influence. It's just not as huge an influence as we would like to believe.

Troy explains how to update to the latest BottomFeeder release on a Mac.

"Redeye" demonstrates that Wes Craven can make a good film without leaving a trail of bodies. We saw this film over the weekend, and I really liked it. There's a fair amount of improbability in the plot - as you watch it, just consider how few options the bad guy really has with the position he's put himself in. Still - the pacing is really good, and you don't think about that as the plot unfolds. Most of the movie is a huge tension building exercise, and I thought it worked.

Rachel McAdams did a great job as the lead - her fear was palpable. Cillian Murphy was nearly perfect as the villain - he had the perfect blend of affable evil down pat. Recommended.

Brad Wilson points to a Dave Thomas post on development:

Dave Thomas has a great blog post talking about language performance. He's talking about Ruby, but his points are universal. It's no secret that I'm in love with dynamic languages in general, and with Ruby in particular. The performance question is one that always comes up early. Ironically, the people asking this question are usually coming from a managed environment like Java or C#, which were subjected to the same questions by C++ junkies. Talking about business apps in general, and web apps in particular, Dave really nails this:

Let's face it: your average commercial application isn't burning CPU cycles solving NP-complete problems. We typically write code that moves chunks of data about and adds up a couple of numbers. In these scenarios, is it worth worrying about the relative performance of the language used to do the moving and adding? Not in my book.

That's something I was getting at the other day in this complexity post - most people aren't writing apps that do huge amounts of number crunching - heck, most people aren't doing apps where anything more involved than a linear search is necessary. And yet, there's an obsession with trivia - like the speed of floating point math. If you'e one of the people for whom that matters, sure - you want a language that optimizes well. Everyone else - you want a language that optimizes at the end that actually matters - developer speed.

Voidstar is out on the extreme edge of Windows opinions, but he makes some good points. Apple does have a real opportunity, if they don't hose us all down by taking the same DRM route that MS looks to be taking. If Apple decides to be consumer friendly, they've got a real shot at increased market share.

I've just released BottomFeeder 4.0 - the relevant upgrades have been moved from "dev" to production, and the download page has been updated. I've also done file releases at SourceForge and Freshmeat - so everything should be up to date. What's new? Here's the list I posted:

1. Tabbed Browsing Support
2. Cleanup of the search engine accesses. External Searches have a single dialog, as do internal searches.
3. Addition of a "Feed Builder Wizard", enabling end users to add search engines as feed builders as they appear
4. Reorganized the toolbar
5. Added full support for Atom 1.0
6. Enhanced the Blog Posting plugin
7. Fixed some bugs in the network layer related to HTTP 1.1
8. Added support for more search engines: IceRocket, DayPop, and Google Blog Search
9. Various and sundry other bug fixes and enhancements

Download the latest and enjoy!

The Croquet project continues to zip along, producing some nifty things:

Croquet is a peer to peer collaborative 3D world. Avatars within the Croquet world can interact with objects, and the lower level messages are replicated within the peer group. He demonstrated this by manipulating some windows in the world, along with more complex objects. He also demonstrated stepping through a window ... a portal ... into a moon/mars scape. Moving through these portals allows the avatars to enter into alternative worlds. They both went through a portal into a "water world" and immediately transformed into fish. As they swam around they came across a "text editor" white board thing ... and edited the text on it.

Ok ... he just opened a paint panel and drew a fish, colored it, and then inflated the 2D fish as he dropped it into the world. Now both of them were able to manipulate and move the new fish around. He entered another portal and showed a interactive spreadsheet, with the inherent ability to graph the values contained in the chart. He also demonstrated using windows as filters to show a filtered view of anything behind it. He was able to move it around showing the wire-frame models beneath the textures mapped onto objects.

All of this is written in Smalltalk, and uses Squeak ... completely cross-platform to Windows, Mac, and Linux. He indicated that they will have a Python, Ruby, and other language support soon.

Hat tip James Governor

Maybe that showdown between free content and the print media that was posited by EPIC is starting. The Times has just placed all of their "name" op/ed columnists behind a pay wall ($49.95 per year). The question I have to ask is, why would anyone pay for that? Forget the political persuasion of any of the writers - there are tons of voices on the net that cover the opinion spectrum. The vast majority of them are free, and can be read quickly and easily with a news aggregator (and the masses will be using syndication, once IE7 ships).

So why would you pay for the privilege of reading the Times' stable of writers? Are they really better than the free pundits? I don't think so, and I think that the Times is in for a rude awakening. The reality is, they just opted out of the political conversation. Up until now, bloggers of all stripes linked to the Times writers, either to agree or disagree. That's not going to happen now - even if a given blogger subscribes, he'll know that most of his readers won't.

I understand that they did this for revenue reasons, but I don't think it's going to work out for them. Unlike the Wall Street Journal, the Times isn't offering unique content. Punditry is just too common now.

Update: Here's a related comment from a political blogger:

There's been much talk about how the Wall Street Journal has actually been able to make money from subscriptions, though I have a hard time believing the Times can duplicate that by selling its opinion pages. According to Michael Wolff, however, while the Journal's internet operations have made money, it's actually hurt their print advertising quite a bit. By putting its internet operations behind the wall the Journal has made itself much less culturally relevant. Why pay for print advertising in the journal when it isn't really part of the buzz anymore?

That's really what the Times is doing - removing itself from the buzz equation. This is the first step down whatever media reorganization path it is that we are heading down.

Tom Yager isn't watching the same industry that I am - he's high on the iTanic:

The road ahead isn’t about 8GHz Xeons or 32-core Opterons. It won’t be about hardware at all. It will be about the $100,000 commercial development suites that will perform automated, distributed build, run, observe, and optimize cycles until native code flows through every possible combination of processor types. And for another $50,000, tools will instrument a commercial app to optimize itself based on changing deployment environments. In this scenario, it might take Microsoft (Profile, Products, Articles) three months just to build a major release of Windows, harnessing the off-hours cycles of every machine on the Redmond campus, but the result would be an OS that could chop a big shop’s system requirements by a third or more.

I’m bullish on IA-64 because a dream world of compilers that take their sweet time to build and optimize but that produce mind-blowing code will surface there first. Everything learned there will transfer to other architectures, however, and we’ll end up with a naturally occurring matrix of CPU types and deployment patterns that provides customers with meaningful choices.

I've had pretty long conversations with our VM team about the IA-64 - they agree that the itanium is just a horrible chip for JITed code. Intel designed this chip back when they thought that C++ was the future of computing. In the succeeding decade, Java and the MS CLR have come to the forefront, affirming the vision that we had in the Smalltalk community a long time ago. More importantly, it demonstrates that the industry isn't moving to anything like what Yager posits. The huge compile farms that Yager speaks of are most certainly not the future of general application development.

Of course, this is the same guy who knew that Apple wasn't moving to intel, so - what should we expect?

PJ Hyett relays this from a JavaLobby thread:

Have you ever tried to go back an make significant changes to a large Smalltalk application that you haven't touched the code for in 3-5 years or maybe didn't even a part in writing? Try that some time if you haven't and then tell us what you think of Smalltalk. If a language can't pass that battle test, it sucks. IMO, Java passes that test very well, much better than Smalltalk or C/C++.

Well, yes. I've picked up Smalltalk code at customer sites that I've either:

• Never seen before
• Have not seen in years

It's never been a big deal to figure out what's going on, even with large codebases that are badly written (and yes, I've seen plenty of large, badly written Smalltalk systems). I spent a number of years as a C programmer previously, and leaving code alone for a week or two would result in a long period of ramp up - along with lots of printouts. I've been using Smalltalk for 13 years now, and I've yet to print code out in order to understand it. Back in the C language family, printing code out in order to eyeball it was a constant experience. Smalltalk, in my experience, is vastly easier to pick back up than C, C++, or Java code. Ruby, I have no idea, but I expect it falls over on the Smalltalk side in terms of ease of pickup.

All you really need is this product, apparently. Napolean should have waited for it :)

Wilkes Joiner thinks that Smalltalk marks out the best path to really getting OO - I like this segment of his post:

Sometimes, I feel the best way to understand OO is to learn Smalltalk. It is such a radically different language and environment that it forces you to think differently. I don't mean to imply that if you've never used Smalltalk then you don't really understand OO. I just think that it can speed up the learning process. My recommendation for anyone new to OO or wanting to sharpen their skills is to download Squeak or VisualWorks, grab a copy of Kent Beck's Smalltalk Best Practice Patterns, and write some code. I've created a very short Squeak tutorial, complete with videos, to get a developer up and running as short of time as possible. There are tons of other resources available out there. Go. Learn.

Warning: The first time I saw Smalltalk was free download of VisualWorks in 2000. I was learning J2EE at the time and my immediate reaction was, "We've been had!" Learning Smalltalk can leave you with a huge level of disappointment with the current "state of the art" languages.

Heh - "We've been had". I like that :)

Danny Ayers is pounding the drum for the semantic web - by which he means RDF. He points to this article by Kendall Clark for backup. meanwhile, Dare Obasanjo has found some contrary voices within the RDF community - Ian Davis and Uche Ogbuji have voiced some doubts about how well RDF actually maps to the semantic web idea. I especially like Ian's take on where the RDF community is:

There are several proposals for dealing with this. The one that seemed to get the most support was to recommend the latter approach and make the first illegal. That means making hundreds of thousands of documents invalid. A second approach was to endorse current practice and change the semantics of the dc:creator term to explictly mean the name of the creator and invent a new term (e.g. creatingEntity) to represent the structured approach.

...

That’s when my crisis struck. I was sitting at the world’s foremost metadata conference in a room full of people who cared deeply about the quality of metadata and we were discussing scraping data from descriptions! Scraping metadata from Dublin Core! I had to go check the dictionary entry for oxymoron just in case that sentence was there! If professional cataloguers are having these kinds of problems with RDF then we are f*****.

That all came in the context of talking about RDF proponent arguments over the proper way to deal with the "author" entity in Dublin Core. Like other groups that have been at it too long, they are down to arguing how many angels can dance on the head of a pin. We may well end up with something like RDF being used for the whole Semantic Web idea - but I doubt that it'll be RDF itself.

I've updated the Silt files on the Silt page - the version you'll find there is for VW 7.3.x. There are new versions of some of the html and css files, as well as the requisite code changes.

I've started dropping cookies for stylesheet selections on the blog - so if you select one of the available sheets, it should now "stick" through the comment and archive pages as well.

There's a large set of developers for whom this statement is really the end-all of their existence:

"If it's not complex, it couldn't possibly scale"

If you want to see an example, look no further than this post from Debu Panda - who argues that you really need the full J2EE stack to build an enterprise website (never mind the complete lack of evidence). Lesscode does an admirable job of taking him down a few notches in the comments - this post in particular is worth reading.

Looks like the tropical systems are getting busy again - the southeast will have to keep a watch on Tropical Depression 18. Looks like we might catch a break with Philippe though.

Well, the "Microsoft in trouble" meme certainly seems to be spreading. Here's BusinessWeek on the subject, and that link popped up in memorandum. I wouldn't cry for them yet - they are sitting on a pile of cash and they are profitable year on year (unlike some other tech firms I can think of). And unlike those other firms, they aren't burning that cash with questionable acquisitions either. I think the "MS is dead" meme is premature, to say the least.

Literally - the Fed is hiring, and Smalltalk is part of the job there :)

Bernard Notarianni explains it all

PR. Differently points to the best gadget I've seen, either for the home office or the road warrior - have a look at this outlet extender. Finally, you can get all the bricks that insist on taking up 2+ spaces in your power strips to play well together.

Update:: Avi points out that the movie showing is actually this one. I was taken in by bad linking from Yahoo's movie page, which linked to the older film - classic data entry error, made bigger by the web :) Check out this page on Yahoo - follow the link to "Cry Wolf" (I expect them to fix it eventually, but it's still bad as I write this). Sadly, the old movie looks better than the new one...

My daughter is off at a long birthday party, so Jackie and I were looking at the movie listings. Imagine my surprise when I ran across "Cry Wolf", showing at all the local multi-plexes - some of them with the number of runs normally allocated to a presumed blockbuster.

Bear in mind that "Cry Wolf" was made in 1947, and stars Errol Flynn! I knew that attendance had been down this summer, but I didn't expect to see 50+ year old movies being resurrected!

Here's a post that mentions a programming oriented boardgame and Squeak - not a combination I would have expected :) The game doesn't look as interesting as RoboRally though :)

Something that hasn't been much commented on in the MS RSS announcements - just how much MS' entry is going to affect the (RSS) tools space. The addition of RSS reading to IE is going to expand the space tremendously, of course - but notice the Outlook RSS capability Scoble mentions? That's going to impact the NewsGator folks quite a bit. Sure, they still have the server side - but I suspect that, over time, their client will end up being a value-added version of whatever MS does. Either way, this entry is going to start consolidating the field of aggregators.

After a jaw dropping number of BottomFeeder downloads last week, things settled back down this week - but still at a rate above what I've been seeing for awhile - 361 per day, on average this week:

Platform	BottomFeeder Downloads
Windows	626
Mac 8/9	376
HPUX	371
Sources	317
Update	313
Mac X	181
Linux x86	179
CE ARM	90
Solaris	22
Windows98/ME	22
AIX	15
Linux Sparc	13
Linux PPC	2
Source Script	1

I'm pleased to see that 98/ME number dropping as well - that's a platform I think we would all like to see go away. So onward - to the HTML page accesses to the blogs:

Tool	Percentage of Accesses
Mozilla	45.6%
Internet Explorer	38%
MSN Bot	7.2%
Google Bot	4.6%
Other	4.6%

The interesting thing about those numbers is how different they are from the aggregator distribution. There's virtually no Mac presence in the HTML numbers, and IE is well represented (although still not a plurality). Things are a lot different in the RSS view - it's something of a different audience:

Tool	Percentage of Accesses
Mozilla	24.2%
BottomFeeder	17.8%
Net News Wire	11.6%
Other	9.4%
Planet Smalltalk	4.6%
Safari RSS	4.2%
Internet Explorer	3.3%
NewsGator	3.1%
BlogLines	2.7%
SharpReader	2.6%
BlogSearch	2.5%
Feed Demon	2.3%
Magpie	2%
Feed Reader	1.9%
RSS Bandit	1.6%
Liferea	1.2%
Feed Tagger	1%
JetBrains	1%
Google Bot	1%
News Fire	1%
MSN Bot	1%

You can see the heavier Mac penetration right there. It's a fascinating divide, and tells you something about the popularity of the Apple platform in the "bleeding edge" crowd.

Dave Winer thinks that the entire tech industry revolves around him. Apparently, he's the Pope, circa 1600 - and gosh forbid anyone mention that the Sun does not, in fact, revolve around him.

So my wife has spent about a week with Outlook (2003) now. Her comment?

They must have spent a lot of quality time with the Ugly Stick

Now, having looked at the video of the new Office 12 - sorry Scoble, but it's not an improvement - in fact, my wife's comments when looking at the new Office 12 were, umm, probably not printable.

Some context - she's new to Office, and the tools she's been using are whatever Sun ships on their Solaris boxes. When you lose a usability race with Unix developers, it's generally a bad sign. There's a ton of wasted space in the UI, simple stuff like "how can I get myself included/not included in a reply to all" - it's hard to find. I last looked at Outlook seriously 5 years ago, and hated it. A lot. This week's experience hasn't made me want to look again.

One of the things I've said we need to work on is deployment. There's work going on to make build up easier than strip down - but in the meantime, the RuntimePackager is getting some long needed improvements. The current version asks you to check the classes you want to keep by namespace or category - most people (myself included!) would rather do that by Package.

So, that's what I've been doing this afternoon - testing the latest RTP build by trying to produce a BottomFeeder runtime in the 7.4 development stream. It's been worth doing - a number of small flaws have been identified, and I think it looks like RTP should be a better tool when VW 7.4 ships.

So everyone and his brother has been raving about the new LINQ stuff in C# and VB.NET - see here for a positive review. I'm sure that developers will use this stuff - and it does look like it could be useful. However, there's a basic approach problem with their solution. They added it by pushing in a bunch of new operators (how many does C# have now?) instead of adding in a library. Adding a library has a lower impact, and it's possible to extend - with a set of operators, you get what the vendor gives you, and that's it. Contrast that with the collection library in Smalltalk: there are four commonly used methods:

• #select:
• #detect:
• #reject:
• #collect:

Those are just methods in the class library though - once I figure out what they do, I can add my own domain specific analogs into the collection libraries, extending my own reach. With the "add a few dozen operators" approach, MS has just made every "learn C# in 21 days" book 50 pages longer.

Yeah, that's what the world needed. Overall, the mainstream continues do damage.

The Register's Andrew Orlowski is continuing his quest to sound important - he's decrying the fact that blogs register so high in general Google searches, and wants them isolated off into their own tab:

Google, along with rival search engines which aped its link based algorithms, has to wrestle with the constantly evolving techniques deployed to trick it into promoting certain web pages. It's an arms race comparable to email spam, and one of the chief culprits is 'blog noise' - a catch all term for the irrelevant blog entries and all the extraneous plumbing that props them up: RSS feeds, empty pages, duplicate pages, TrackBacks, and so on.

Yeah, that's it - we should only trust the "mainstream media" - people like Orlowski, who just make crap up? Like the NY Times, who had Jayson Blair? Like CBS News last year, with their memos? Doesn't seem to me that the vaunted editorial staffs have done any better than the "irrelevant" blogs. I think what Orlowski really objects to is that - with blog results showing so prominently in Google - we can find out just how much of a hack he really is.

Well, the excitement never ends with a Windows box. My wife's Media Center TV shipped with no TV out, so we had to buy a new video card. That's not what I'm this is about though - HP decided what hardware to include, not Microsoft. I ran down to CompUSA and picked up a Radeon card (not a top of the line one - basically the same as the built in support, but with TV out). That's when the excitement started.

We tossed in the new card, flipped the VGA cable from the old (built in) port to the new port. Reboot, the XP screen shows.... and then winks out. First check - Windows supports two screens, but doesn't check the ports, apparently. So, we have to flip the cables back, tell Windows to use the secondary screen as the primary, flip the cables again, and have it work. Plug and play, it's not.

Next, Media Center stopped working. Ok.... I understand needing an updated driver. But the message we got was "Some files that are needed aren't installed. Reboot or install". Yeah, that was helpful. So the flat panel works again (we're back where we started!). Now, we tried hooking the composite video (RCA cable) from the TV out port to the VCR. Then, we tried playing some content. Nothing - Displayed just fine on the monitor, but nothing out to the TV.

Great. Now I'm back to reading manuals and websites. I keep hearing people rave about Media Center. So my wife fiddles with the thing for awhile, finds some help that directs her to the TV out settings for the card. First off, she gets a black and white bounce screen - like an old TV that has rabbit ears, it looks like. To get that far, we made the mistake of telling it that the TV was the primary out - at which point Media Center pretty much refused to accept settings until we turned full screen off, which required getting the mouse "just so" on the bar. There's a UI guy in Redmond who should be beaten over this

After getting that, we dove into the Catalyst configuration screen for the Radeon card. Oh boy - this is the UI from hell - the ATI guys need to visit the hall of shame. Fixed sized dialog, and if you don't set your fonts to the "too $%^&*( small to read" setting, it doesn't show all the information. Swell. After a lot of mucking around, we got the video scroll fixed - and we aren't entirely certain how.

I have news for Scoble and the rest of the MS gang - Media Center PC's may be selling like hotcakes, but that doesn't mean that average people are going to go through the sheer hell of setting up TV out. You want to know what you're competing against? Buy a TiVo or ReplayTV. You plug audio and video cables into the obvious in/out slots, and then you go mess with the on screen programming. There's a lot less swearing involved, and a lot fewer hostile defaults to be overcome. Right now, Media Center PC's are great if you want to watch TV on your PC, but they suck eggs if you want to push video out to a TV.

One of the things that you learn when you have a blog is the permanence of things that you've written. You also learn that what you think is important may not mesh with what other people think. Where am I going with this?

Well, I wander through my referer list every day, and - other than adding spam sites to the blacklist - notice which posts keep getting referenced. Here's an example - I posted on the illogical pride that people take in being innumerate awhile back, and now I see a referal from innumeracy.com just about every day. Which I find interesting, because it's not a topic I write about all that often. I still feel strongly about that, but it's not something I focus on... and yet, there it is.

And that's one of the things you learn as you continue to roll posts out over time. It's a permanent record of sorts - even if the urls change or the site goes down, those posts exist in various caches (Google, Feedster, etc) - and will be there for a long time. Which brings something to mind - young people who've started a blog may well find posts coming back to haunt them decades later. Just think about the current Supreme Court hearings on John Roberts (regardless of where you stand on it) - there's been a scramble for everything he's ever written, in an attempt to parse his future thinking. Now, imagine it's 20 years from now, and a person in the same situation had a blog.

It reminds me of a book I read a few years ago, "The Truth Machine". Part of the scenario laid out in that book was archival recording of just about every moment of daily life. It had some fascinating consequences, and I'd recommend the book highly. To a limited extent, the archived posts being put up by so many of us are becoming that scenario.

Thought Tracker makes a good point about simplicity and productivity:

The bigger the language (syn)tax is, the bigger will be the distance to the Domain. (read: you will have more syntax-noise in the code). This "noise" can help you to understand the tehnology, but it troubles you in understanding the domain. That's why Smalltalk was good: very simple syntax. The rest was/is Domain Language.

This is one of the reasons I consistently harp on the complexity of Java and C# - they add capability by larding on more syntax, due to the inherent weakness of their original designs. Generics? In Smalltalk, we don't really have to think too deeply about the problem - unlimited polymorphism and DNU handlers gives us all we need, without having to learn a bunch of extra language rules. The Java guys spent years pondering the question, and then larded on syntax. MS didn't spend the same amount of ponder time, but they also larded on syntax.

Every time that happens, it gets harder to solve the actual problem at hand - because you have to learn the extra stuff they layered on instead of just diving into it.

The Smalltalk User Community in Boston has kick started itself - Colin Putney has an update:

I'm happy to announce that after the success of our first meeting a while back we're going to holding monthly meetings of the Boston Area Smalltalk User's Group on the third Tuesday of each month. This month's meeting will be held in Somerville.

Date: Tuesday, September 20, 2005

Time: 6:00 pm

The Joshua Tree Bar & Grille

256 Elm St.

Somerville, MA 02144

617-623-9910

Map

Check it out if you're in the area

ESUG is working to get more Smalltalk content out there - they are paying 100 Euros for each qualifying item. Have a look at the linked page for details.

In the "stranger than strange" category, comes this story: Inventor fuels car with dead cats

he Web site of Koch's firm, "Alphakat GmbH", says his patented "KDV 500" machine can produce what he calls the "bio-diesel" fuel at about 23 euro cents (30 cents) a liter, which is about one-fifth the price at petrol stations now.

Koch said around 20 dead cats added into the mix could help produce enough fuel to fill up a 50-liter (11 gallon) tank.

JenSense reports that Yahoo is adding contextual ads to their RSS - InfoWorld has been doing something like this for awhile now. Anyone who's shocked or irritated needs to go beat themselves with a cluestick :)

I've got the new development builds up - visit the BottomFeeder download page, and grab the appropriate build. If you already have it installed, just get the appropriate baseapp-*.zip file, and decompress it in the install directory. Remember to scroll down to the "dev" links first.

This is a candidate 4.0 build - if it holds up (I'm running it now), I'll formally release it shortly.

ThreadWatch has a good explanation of what the new Google search is, and what it isn't:

What it is, is feed search. On the surface there's not a greatly noticeable difference for many, but the difference is there, and it's quite profound once you understand why.

Many feed items are truncated, or otherwise edited - just snippets of the complete "on site" post. This of course means that you're not getting the full picture. You're getting a pretty good slice of that picture, but it's not complete, not by a long shot.

So does that give full text feeds an advantage? I would think it does. Given Google's dominance in the search space, this fact all by itself may change the dynamics on full/partial feeds. There's going to be a cross current - people and companies want traffic to come to the site (not least for the ads), thus driving the decision toward partial feed content. On the other hand, here's Google, effectively cutting those sites off at the knees in terms of search rank.

This should make things interesting, to say the least.

I think BottomFeeder 4.0 is just about done. Rich has been an immense help in the areas of bugs and consistency - the release will be a lot better due to his help. The docs should be arriving in my email soon, and then I'll do a putative build. If that runs ok for a few days, I'll push it out the door.

After the long winded discussion of refactoring here, Alan added some light to the heat over here. Today, Peter William Lount adds some more:

Those that think that types are not used in Smalltalk are not entirely correct, "types" are used, or more correctly the objects class meta information is available for use at runtime to perform any necessary and all possible 'type' operations at run-time (which includes almost all the compile time "type" possibilities). The object meta data available through the base Smalltalk language while the program is running is a much more powerful facility and provides capabilities not possible in staticly compiled typed systems such as C, Java, C++, C#, ... . These and other statically compiled typed language and systems strip away most if not all of the "type meta data" so that your program is a barren husk devoid of the richness that runtime meta data provides.

Read the whole thing.

InfoWorld has a fascinating story on a build vs. buy scenario at a railroad:

Around this time we sent a few IT troopers down to KS to see what was going on. It was the last point the project could have been killed or redirected, so I checked in with my friend Michael, a savvy project-management contractor. He did an analysis comparing the functionality and cost of reinventing the system in house with modifying the business processes involved in the KS model; KS's business processes were very different from ours. Michael calculated that if we continued with the collaboration as planned the price would not be $30 million or even $60 million. It would be closer to $70 million. The cost of developing a new system from scratch -- one that would perform the same functions as the KS system and be properly designed for our business groups -- would only run about $45 million.

Of course, management refused to budge from the existing project, even as costs escalated. Once the numbers get "too big", politics comes into play - no one wants to be identified with a failure on a large scale, so an "all is well" attitude takes over.

The basic problem is the premise - that "buy" is always better than "build".Often it is - especially with commodity software. The trouble is spotting which parts of your businesses can't easily be handled by commodity software, and dealing with those. Many companies have wasted millions on ERP installations over this. Why? Because a pre-built ERP system will impose a set of rules on how operations will run. Those rules may well be fine, in the abstract. What they don't account for is the peculiar culture of your company. Ask yourself - as hard as it is to change software, is it actually easier to change corporate culture?

Scoble has a video interview up on the new Office 12 - as has been reported elsewhere, the UI is very different. I started out very skeptical on the "no menu" approach, but - as the video progressed - it looked better.

Having said that - my main usability issue with Word remains unanswered - will a bullet point go where I want it, rather than where Word wants it? That's the one thing that makes me despise the entire product.

Google has launched a blog specific search site - but where are the results in RSS or Atom?

Update: Scoble is marginally more upbeat on this. He also noticed what I missed - there are results in RSS and Atom at the bottom of the page, ready for subscription.

I've put the dev stream update and a new dev build out for BottomFeeder. Now, a caveat - I had trouble using the update, so I created the fresh build. If you grab the update and run into problems, try restarting the app. If that doesn't address the problem, then download:

• This file for Windows
• This file for Windows98/ME
• This file for all other platforms

And unzip the file in the BottomFeeder directory. Delete all the files in the "app" subdirectory, and then restart.

Update: Grab the update to the BottomFeeder parcel, and then restart. It will all work fine after that.

This story notes that Delta and Northwest are about to head into Bankruptcy protection - while United and USAirways are still there. This begs the question - of the old line "major" carriers - United, American, Northwest, Delta, USAirways, Continental, how many will actually survive? Deregulation and airlines like Southwest have changed the rules in the airline business, and the surviving majors have adapted badly. I suspect that we'll see a lot of consolidation over the next 3 years - I wouldn't be at all surprised to see only 2-3 of the six I mentioned left standing after the dust settles.

I've got another dev stream update out for BottomFeeder (I'm talking about the online updates, not the full builds) with more tab enhancements. Rich pointed out that internal searches should really open into a tab, which makes sense - so now they do. He also pointed out numerous small inconsistencies in the menus, which are now corrected. It's getting closer to 4.0

Update: I pulled the update, due to some problems I noticed in the runtime with this update. Stay tuned.

This Salon article is about a college anthropology professor who felt so disconnected from her students that she went undercover to study them. She came to what I'd have to call "well, duh" conclusions:

There is an awful lot of conversation about nonacademic, nonpolitical, nonphilosophical things, but I saw something very interesting also. Anyone who said they did have a philosophical conversation might qualify it, like, "Yeah, we were really drunk that night, so we got into all this deep philosophical stuff," or "Yeah, sometimes I get into this dorky mood and then I talk about deep topics." When you hear that as an anthropologist, you think the students are responding to a criticism that isn't even being made, that is in their head.

What do you mean? What's the criticism?

Students don't like to sound like they're trying too hard. That's what I would see in the pre-class conversations. You know, "How'd you do?" "Pretty good. I got an A, but I barely studied." Or "I did well, but it's amazing, because I thought I totally bombed this." You have to seem like [success] is effortless, or like you haven't put a lot of work into it. And that becomes part of the culture. I think a lot of students want to have [more substantial conversations], but they don't feel comfortable doing so.

Where did she go to school in the first place, and what planet were she and her fellow students on at the time? The scenario she describes sounds an awful lot like my college years at SUNY Albany, 1980-1984. Politics came up rarely, no one wanted to look like they were working hard, there was a lot of partying - I rather suspect that schools have been like this for a long time (with varying levels of open-ness about it over time). Heck, even in the supposedly political 60's, I would bet good money that most students spent more time ogling girls (or boys) than they did protesting.

Maybe it's not so much that the students have changed - and more that the professor hung around with an out of the mainstream group in college - and thus defined that experience as the norm.

I haven't commented on the Ebay acquisition of Skype, because I haven't given it a lot of thought - I don't follow Ebay at all. And while I use Skype, it's not something I depend on. Here's my first thought on it though: Why?

Ebay does one thing - internet auctions. Exactly how is that related to a telecom offshoot like Skype? This smells like the all too common "money was burning a hole in management's pocket" problem. I saw a lot of that (on a much smaller scale) at ParcPlace-Digitalk - management went on a buying spree, mostly because they could. I see no synergy here - it's a buy into a business that's going to require focus and investment (there's going to be a lot of interaction with regulatory groups in the medium term - 911 service, etc, etc). Will Ebay want to invest that level of management focus? How exactly does it benefit them?

There's another automated aggregation service out there - memeorandum. Scoble was finally able to break the news as it launched - he was given early access, apparently. Right now it's doing aggregation in two areas:

• Technology news
• Politics

I just subscribed to the tech news service in BottomFeeder, and I'll have to see how it works out. One thing that's going to limit my interest - the aggregated items are extracts, not full items. That makes sense, as ripping content from someone else's site via a robot is anti-social - but on the other hand, it means extra steps for me as an end user. I'll see what I think after a few weeks of usage.

Loose Wire spots a potential problem for the future in the (apparently botched) iTunes 5.0 update - the lack of music in a physical form. Many of us are buying songs online now, and - once the roipping of our existing collection is done - just not buying cd's anymore. Well, what about hard drive crashes, malware, and botched updates? Loose Wire comments:

This kind of thing scares me. It scares me because we don’t yet grasp how fragile our music collection has become. Before we had a pile of CDs we could always go back to if our tapes, MP3s or burned CDs gave up the ghost. Nowadays our music collection may be just in the form of MP3 files, and what happens to them if something goes wrong? What happens if MP3 software (or a system crash, a hard drive error, or a stray catheter) corrupts your files, your tags, or your authorisation and proof of purchase? At what point do we say, “forget this, I’m not going to pay for anything that doesn’t come in some physical form I can stash on a shelf”?

That's a good question. I doubt you'll see mass moves in that direction, but I do expect to start seeing curmudgeons.

Scoble reports on the coolness of the iPod nano - people at the MS PDC are buzzing about it. Apple's marketing is firing on all cylinders - the nano almost makes me wish I hadn't bought an iPod mini. Heck, I'm considering buying one anyway, and sliding the mini off to my daughter.

Back to Microsoft - there's nothing they make that inspires that kind of desire in the marketplace. I don't know if that's actually a problem - large market share and profits are kind of a reward all by themselves :)

This is the sort of thinking that irritates me - in an article covering the increasing ability we have to customize what we watch and listen to, a researcher decries the personalization:

"I think that ends up creating a cultural divide among us, because we don't have as many common experiences. And when it comes to the news of the day, even the most important news, you can't assume that any of your friends or neighbors watched the newscast and know what it is," McCall said. "It means there's going to be less opportunity for generations to understand each other, and people to understand each other socioeconomically."

Translation: "It was a lot better when smarter people selected what you needed to see and hear". Yeah, that's what I want to go back to - a choice of three channels, whether I like them or not. Thanks, but no thanks.

I've spent the day applying fixes and changes to a bunch of changes to various codebases. First, Rich continues to send me things that need fixing or enhancement - I've added a lot of stuff over the last few releases, and Rich has spotted various issues with the tools. That should make the release better.

I also added user settable styles to the blog pages today - you can click one of the available style links, and see which ones you like. Finally, there's apparently been a problem with the NC download app for a few days - I managed to not update a method at some point, and you couldn't progress from the login page to the downloads - very bad. That's fixed now - sorry about that.

But heck - I should look at the bright side - at least no one around here messed up the power connections, dropping 2 million people from the world of electricity (as happened in Los Angeles this afternoon).

The blogs have all been updated to support user settable styles - check the list of available ones on the link bar, and see which one you like!

How to avoid tracks:

The Salt Lake Tribune publishes a tip for aspiring politicians from Utah Gov. Jon Huntsman (R): "I don't do e-mail because I think you can always be set up for an unfortunate incident if you spill your soul."

Instead of sending e-mail, Huntsman pens notecards. The article notes "the quaint practice of writing on notecards also apparently allows the governor to sidestep the state's open-records law. Requests to see some of his handwritten notes have been thwarted."

Now, translate that from the political to the business world via Sarbanes-Oxley - you think C level staff aren't looking for similar ways to thwart that? Sometimes, the desire to "see everything" actually results in seeing less. I'm not sure where the right balance point is, but I'm thinking that we may have drifted away from it.

Well, this is certainly curious. Over on the Instantiations site, it looks like they've done away with the non-commercial download of VA. Have a look at the license agreement terms:

EVALUATION. Instantiations grants you a nonexclusive, nontransferable license to use the Software. You may 1) use the Software only for internal evaluation, testing or demonstration purposes, on a trial or "try-and-buy" basis and 2) make and install a reasonable number of copies of the Software in support of such use. The terms of this license apply to each copy you make. You will reproduce the copyright notice and any other legends of ownership on each copy, or partial copy, of the Software. You will 1) maintain a record of all copies of the Software and 2) ensure that anyone who uses the Software does so only for your authorized use and in compliance with the terms of this Agreement. This license begins with your first use of the Software and ends 1) as of the duration or date specified in the documentation accompanying the Software or 2) when the Software automatically disables itself. Unless Instantiations specifies in the documentation accompanying the Software that you may retain the Software (in which case, an additional charge may apply), you will destroy the Software and all copies made of it within ten days of when this license ends.

Time bombed evals, how very 1990's of them. I could be wrong about this; it's not completely clear to me whether the evals are time bombed or not (anyone there care to clarify?), or what the term is if they are. Time bombed evals are a really bad idea, which is why we don't use them. It takes time to evaluate something as large as a set of development and deployment tools - we've had more than one customer who's evaluation term lasted from the time of download until the point of deployment - at which time they bought an appropriate license from us.

If we had used time bombed evals, we would have lost those sales - there's no doubt about it. So, if you want to do a full evaluation of a professional Smalltalk environment, there's now only one place to go - here.

Ok, this will seem silly, but I was very pleased just now when I stumbled across a setting in iTunes and the iPod - the one that equalizes the output for all songs. The fact that songs from my older CD's were quieter than the ones from newer CD's and purchased music was driving me nuts :)

With the new tab support in BottomFeeder, I thought I should explain one of the customizations - the tab menus. As it happens, the tab widget in VisualWorks doesn't support menus out of the box. So how did I add them? Well, first off, I noticed that Typeless has tab menus (Typeless also ships as a BottomFeeder plugin, which is how I use it). So, I loaded TL from the Public Store and started having a look around. It turns out that Michael extended the widget in the most natural fashion, by creating a new controller subclass. Normally, the tab bar is managed by a class called TabControlBarController. So, first thing - he created this class:

Smalltalk.UI defineClass: #TabControlBarControllerWithMenu
	superclass: #{UI.TabControlBarController}
	indexedType: #none
	private: false
	instanceVariableNames: 'menuHolder performer owner mouseDownIndex '
	classInstanceVariableNames: ''
	imports: ''
	category: 'UIBasics-Controllers'

Then you have to modify the view class, so that it uses this controller (or, at the time you create the view, replace it. This code just replaces the default controller) - here's TabControlBarView>>defaultControllerClass:

defaultControllerClass
	^TabControlBarControllerWithMenu

That makes sure that we always get menu support when we create a tab widget - assuming that the new controller class is implemented properly. Here's what's going on there:

• First, add some new instance variables to handle menus and the state required for them: menuHolder performer owner mouseDownIndex
• implement #yellowButtonPressedEvent: so as to know when to pop the menu
• implement all the other supporting methods

There's a small side story in the name of #yellowButtonPressedEvent: - back in the day, at Xerox PARC, the mouse actually had colored buttons - red, yellow, and blue. Those names still exist in the code for VW (and, I think, for Squeak). Anyway - the method:

yellowButtonPressedEvent: event 
	"we about to bring up a menu, unlock the event
	queue so that we can process expose events."
	| index |
	view numberOfElements = 0 ifTrue: [^self]. 
	index := self findElementFor: (self sensor cursorPointFor: event).
	(self owner respondsTo: #adjustTabbarMenuFor:) ifTrue:
		[self owner adjustTabbarMenuFor: index].

	self sensor windowSensor queueLocked: false.
	self processMenuAt: event globalPoint centered: true.
	^nil

The reference to #adjustTabbarMenuFor: is something I missed out of the gate. I had to implement that in my class, in order to ensure that the menus were set up correctly - it looks like this:

adjustTabbarMenuFor: anIndex 
	| tabModel sub tabbed menu |
	anIndex = 0 ifTrue: [^self].
	tabModel := self browserTabs at: anIndex.
	sub := self widgetAt: #feedID.
	tabbed := self getComponentFromSubcanvas: sub withID: #TwoflowerHTML1.
	menu := self class tabMenu.
	(tabbed widget tabBar component controller)
		menuHolder: (ValueHolder with: menu);
		performer: tabModel;
		owner: self.
	self checkTabMenuEnablement: menu

All of which does the following - make sure that a new tab has a menu, and that it's attached to that menu. As well, make sure that menu items are in the right enablement state. There's a few other setup methods, but you can see all of that by browsing the package RSSExtensions in the Public Store. The next important thing was setting the tabs up properly in my UI class. In my #postOpenWith: method (which fires after the UI opens, I added the following:

self presetTabs.
tab := self changedTab.
self setupTabMenu: tab

presetTabs
	| browserTab |
	browserTabs := OrderedCollection new.
	browserTab := RSSZoomItem new.
	browserTabs add: browserTab.
	self tabs list add: browserTab displayString.
	self tabs selectionIndex: 1.
	self setupEventHandlingForTab: browserTab

changedTab
	"changed to a new tab; adjust"

	| index  browserTab sub tabbed |
	index := self tabs selectionIndex max: 1.
	browserTab := self browserTabs at: index.
	sub := self widgetAt: #feedID.
	tabbed := self getComponentFromSubcanvas: sub withID: #TwoflowerHTML1.
	tabbed widget client: self htmlModel spec: #windowSpec builder: self builder.
	(browserTab feed isNil or: [browserTab feed isFake])
		ifFalse: [self focusOnItem: browserTab item].
	self tabs selectionIndex = 0
		ifTrue: [self tabs selectionIndex: 1].
	self setCurrentTabDetails.
	self setupHTMLPane.
	self setBrowserEvents.
	self restoreHistoryFrom: browserTab.
	^browserTab

setupTabMenu: aZoomItem
	| tabbed sub menu |
	sub := self widgetAt: #feedID.
	tabbed := self getComponentFromSubcanvas: sub withID: #TwoflowerHTML1.
	menu := self class tabMenu.
	(tabbed widget tabBar component controller)
		menuHolder: (ValueHolder with: menu);
		performer: aZoomItem;
		owner: self.
	self checkTabMenuEnablement: menu

Those three methods are the heart of the support. The first one sets the tabs up - with the proper domain model (in this case, an object that holds the selected feed, the selected item, and the history), and sets the current tab index. The second method, #changedTab, fires whenever a tab is selected. It ensures that the correct stuff gets displayed, and that state stays correct. The last one, #setupTabMenu: makes sure that menus are set up for selected tabs.

There's some infrastructure there - event handling, and some application specific stuff - but that's the gist of it. If you need help doing something like this, just send me an email.

I think I like the "Fire" style best. I left the "Water" one up for a day, and just switched back to the "Fire" pattern now. Anyone have an opinion?

Believe it or not, Apple has them too. There is no silver bullet in the current platform choices out there.