Smalltalk Tidbits, Industry Rants

Via Lambda the Ultimate comes a reference to a talk given by Dan Ingalls on Smalltalk in 2001:

The nice thing about a language that takes hold is that you can work with it again and again. In 30 years we have built Smalltalk systems with quite different constraints. This talk will examine a few of these, and show how tricks of the trade can be applied to enhance one aspect or another and, frequently, to make real progress.

There are video feeds of the talk - check it out.

Smalltalk Solutions 2003 Would like to announce its technical conference schedule.

Links of Interest:

• To view this year's presentation schedule please go to the presentations page
• Also, we would like to announce the tutorial schedule. To view the tutorial schedule, please visit the tutorials page
• To register for the show, please visit the Smalltalk Solutions Home Page

Don't forget, early registration ends June 13th, 2003.

See you all in Toronto!

For additional information please contact: Joy Murray, Conference Coordinator

We had probably the bext day in 3 weeks yesterday - a few minutes of sunshine in between clouds. So I was watching the fantasy (I mean weather) report last night, and they told us clouds and sun, followed by rain again Wednesday. Better than we've had, even if it stinks... so I get up today to slate gray skies, and sure enough - rain by noon.

A few days is one thing, but 3 weeks? It sucks the energy right out of you.....

Bob Lewis tells us things we don't want to hear:

Think of what you have to offer an employer as your product, and what you do when you look for a job or try to keep the one you have as marketing and selling that product, and I think this will make the point clear. If not ... imagine you're Lucent and you have to compete with other PBX providers in an incredibly tight market. You have to build all of your costs into your pricing. You'll buy your chips from the lowest cost provider that delivers to your specifications; likewise engineering and programming. What else can you do - charge a lot more than your competitors and try to make it up in "brand management"?

The fact of the matter is that American technical professionals don't want to compete on price with their Asian and Russian counterparts. That's fine, but if we don't, we'd better find some other dimension in which we can compete successfully, because there's nothing in the social contract that obliges our customers ... the companies that employ us ... to pay more for the same service than they have to in the labor marketplace.

In fact, I think it would be pretty easy to construct an ethical argument that if American programmers charge more than their Asian counterparts for the same services, then it's the American programmers who are guilty of greed, not their employers. It comes down to how you define "fair price," I guess.

Ouch. Hard to argue with though.

Over at IUnknown:

I wonder if it's just me, or whether the community that I frequent has this on its collective consciousness, but I've been spending quite a bit of time wondering about the benefits of dynamically typed languages.

I take this as a very good sign. And I liked this:

Please do not mistake static/dynamic typing for strong/weak typing. Ruby is actually a strong dynamic typed language

Not everyone is as confused on this point as Ted Neward clearly is.

Via Dave Buck I got pointed to this article by Jim Waldo - apparently Sun's lead guy on Jini. In discussing standards, he says this:

What gets lost in all of this, of course, is whether the technology being blessed by the standard actually helps the customer to solve a real problem. Can the standard be implemented? Will the implementation perform adequately? How much will it cost? All of these sorts of questions are not appropriate in the era of de jure standards.

What is also forgotten in all of this is how fragile the de jure standards have been in the past. I can't think of a single standard that was invented by committee that has survived in the marketplace. The long-standing standards are those that were first de facto standards, and were described (no invented) by the standards bodies.

Such standards didn't start out in a standards body. They started out solving problems. Because they solved the problems, people used them. The use drove the standard, not the other way around. This allows innovation, this allows technical progress. Things that work get used by people who are trying to solve problems.

This does take the decision making power out of the hands of the managers, and the IT departments, and the technical analysts. They aren't trying to solve problems in new ways; they are trying to lead parades, or keep their jobs, or show that they have influence. They aren't the engineers that can actually understand the solutions, but they do (for the most part) understand politics. Standards groups cater to their expertise, not the expertise of the engineer.

Of course, this hard dichotomy is something of a caricature. There are substantive discussion of technical merit in standards groups that are trying to invent. But that isn't all that goes on. And certainly there is little evidence that the best technology wins in such groups. Just look around you for evidence of that.

So the next time you are talking to a manager and he or she tells you that you have to use something "because it is a standard", push back. Ask why only standards can be used. Ask if the standard has actually been implemented, or if the standard will really solve the problem under discussion. For that matter, ask if the manager really knows what the standard is. If any of these questions can't be clearly answered, maybe the standard isn't the way you should approach your problem.

Hmmm. J2EE comes to mind. A big mass of code, designed by people who had (apparently) never written a distributed application, with no actual customer feedback. So - based on one of Sun's senior staff's recommendations - you should reject efforts to implement J2EE. When management balks, quote the Sun guy!

Via Sax.net comes this interesting twist:

And the saga continues... now Novell has released a press release, claiming that they, not SCO, own the patents and copyright to the "open source" Linux code in question.  The press release also contains a letter Novell wrote to SCO:

Importantly, and contrary to SCO's assertions, SCO is not the owner of the UNIX copyrights. Not only would a quick check of U.S. Copyright Office records reveal this fact, but a review of the asset transfer agreement between Novell and SCO confirms it.

Maybe someone should buy the movie rights....

I spotted this in comp.lang.java.advocacy yesterday:

I work for a large company (20,000 employees) that is planning the re-write of a portion of its home-grown 2-tier ERP into a multi-tier J2EE environment. The bulk of the work will be outsourced, with future maintenance and enhancement done in-house. The application has about 1000 users in NY, with about 100-200 concurrent transactions during peak periods

Not replaced by an off the shlf package - a rewrite. I would have thought that this kind of insanity was past - what these people are going to do is recreate a bunch of stuff they already have - for large amounts of money. I posted a query as to why they were going this way, and what the ROI would be - but no answer. And people wonder why business units have no respect for IS....

I grabbed the old OLE code - the stuff that was being done for the Van Gogh release of VW (the one that would have had native widgets, the then current version of Store, and OLE). One of the many casualties of the PPS/Digitalk merger. Anyway, I grabbed the code, got it loaded, and published it to the public store. You'll want to visit this page for information and pointers to the original .st files - the ones I used to create the OLE bundle. The code loads into an image and can be published back out of an image - beyond that, I can't guarantee anything. COM/OLE is something I know very little about. Back in the day, I used the pre-releases of the this code to demonstrate Excel embedded in a VW 2.0 window - so all you C/Smalltalk hackers, have a look.

Silvermark and Cincom have agreed to deliver an evaluation version of TestMentor for VisualWorks with each release of VisualWorks. Additionally, Cincom will be reselling the TtestMentor product, and the two companies will cooperate on consulting opportunities. TestMentor will start shipping with VisualWorks 7.2 - both commercial and non-commercial.

I have to get up at an insanely early hour for my trip to Ottawa tomorrow morning. I'm actually flying into Montreal - it's cheaper, go figure. I then drivee to a lunch meeting and a customer visit, followed by the XP talk. Busy day ahead.

So here I am in the plane, on the way to Montreal - yeah, Montreal, from where I'll drive to Ottawa. I have to go through New York though, on a lovely little prop plane - I love it when I can't stand up in the aisle without hitting my head (it's not as if I'm tall). Oh well - I have to shut down the PC as we arrive at LGA....

The Fishbowl has some interesting points about a topic that many of us would prefer to avoid - the nasty bits of code we have to write to get around limitations in the frameworks, libraries, and languages we use:

You are in some way subject to architectural, framework or language constraints that force you to write ugly code. For example, your UI framework requires one kind of object, your persistence framework requires another, and you keep having to convert between the two.

His solution?

Hide your ugly code inside a Ghetto. The ghetto is a single file or class where issues of code cleanliness do not apply. It is entered by reputable developers with no small amount of trepidation, and left as quickly as possible. On the other hand, it does the job, and it keeps the bad elements away from more cultured code.

Yes, I've had to do this with BottomFeeder. I think everyone has. WhatIreally like is his Examples of Use:

None that this author is willing to admit to.

Heh. Yeah, I get that :)

Those of you not using 7 or 7.1 yet (upgrade!) should look here - we have updated engines, which may be used with VW 5i.2 and up.

I'm sitting here in the Ottawa airport, at a desk with a phone that has a jack. My computer can dial the phone; it even finds the modem on the far end. But after each connection, it just drops. But wait - it was the phone! A connection, happiness....

The meeting went pretty well - it was a friendly crowd - I was told that some of the Java advocates didn't show for the meeting. Maybe having Dave Astels talk last month, followed by me this month was too much Smalltalk for them - one way or the other. In any event, it was a lot like the DC talk I gave awhile back - lots of give and take with the audience, lots of good questions. My thanks to David Buck for setting it up.

Earlier in the day, I had a meeting with a customer in the Ottawa area. They are mostly still on VW 3 - they have been avoiding moving from ENVY. However, a lot of what we have done in VW - especially the network clients work since 5i.3 - is stuff they really, really need. I gave my stock What's next in Cincom Smalltalk talk - they were pleased with what they saw of the new tools, and pleased by what they saw in terms of upgraded functionality. A lot of good questions, with answers that made them happy - a good visit.

After the XP meeting, there was another customer visit this AM. I met with an account that Dave Buck has been helping - they have migrated, with his help, from 2.5.2 to 5i.4 (they plan to go to 7.1, but are in the middle of their development cycle). They liked the look of the RB and the new debugger, and specially liked the RB extensions for SUnit - Dave Buck has made them into testing believers. This was a good meeting as well - and it ended on a very positive note - we met a woman who had only just learnd Smalltalk - shee's a domain expert at this outfit, with a CS background - but had not done development since college. Her statement to us:

Smalltalk was so asy to learn. Dave (Buck) helpd me along, but it was so simple that I was able to pick it up within a few weeks

Not an exact quote; I'm going from memory here. I constantly hear from people that "they can't find Smalltalkers". They aren't hard to train - it's a simple language!

One more thing - I want to thank David Mattinson of Software Mechanics - without his help I never would have found my hotel last night! If you need Software Development help in Ottawa area, look him up - he's a great guy.

I've posted about Dynamic Typing a lot recently, but here's a great post I just came across:

From the Walnut guide to the E language

"Many of the most complex yet most reliable systems in the world today have been developed with dynamically typed languages. If you are a Java programmer, unshakeably convinced of the perfect correctness of static typing, all we can do is urge you to try E first and form your conclusions later. We believe you will find the experience both pleasant and productive, as the long heritage of programmers from Scheme to Smalltalk to Perl and Python have found in the past."

There you go - it's nice to have allies.

Have a look at http://wiki.cs.uiuc.edu/VisualWorks/Smalltalk-80+in+a+box - tomorrow, 10:58 am Pacific Time marks 20 years since Smalltalk-80 Version 2 image was saved. Version 2 was the general release version of ST-80. Toasts all around!

Well - AOL dropped the suit and MS paid a bunch of cash - and now we see that AOL will be keeping IE as their browser. So given this:

• AOL will use IE
• The AOL suit against MS is now history, in AOL's favor

What's the liklihood that AOL will contribute anything to the Mozilla codebase any longer? The Mozilla project is huge - it's not going to move forward much without some backing...

Via Jeff Zeldman comes word of a nifty sounding cleaner for those Word docs that you have to post as HTML:

"We must put all our Word documents online." Eight words that strike fear into the hearts of web professionals everywhere.

Microsoft Word is the default tool of businesses and organizations, and it includes a button to convert documents into HTML. But the HTML Word generates is littered with invalid, proprietary tags and attributes that are included, not to facilitate web publishing, but to ensure that Word docs and Word-generated HTML cannot be edited by non-Microsoft programs and will not display correctly in non-Microsoft browsers. It is the bane of web agencies, freelancers, and in-house professionals charged with the maintenance of large content sites with low budgets.

Textism's Word HTML Cleaner cleans up the junk HTML that Microsoft Word generates, removing proprietary crud while leaving basic formatting and typographic entities (like curly quotes) intact. Created by Textism's Dean Allen, the online tool is available for your use absolutely free. This product is a life saver, and although it is offered free of charge, you might want to slip a buck or two into Mr Allen's virtual tip jar.

I'll have to have a look at that.

I've posted on the SCO mess here and here. Now, Frank Hayes of ComputerWorld has the most cogent summary I've seen so far:

So what do you do if you're the CEO of a $65 million software company that's losing money, losing market share and -- worst of all -- has a stock that's lost 99% of its value in the space of two years? If you're Darl McBride of The SCO Group, you file a billion-dollar lawsuit against IBM. Then drop the product you've staked your future on. Then send threatening letters to about 1,500 of your biggest potential customers. Then announce a deal with Microsoft. Result: SCO's stock is up 500% since January. Sort of takes the mystery out of why SCO is taking wild swings at the Linux it championed just months ago, doesn't it?

From the Smalltalk IRC Channel:

"Developers are like simple chemicals (no, not because they're cheap and smelly)"

"In the right combination, they can make miracles.  Screw up the mix, and you can blow up a city"

Dan Gillmor writes:

Denise Howell has posted Bill Gates' and Steve Jobs' remarks on her weblog. Should I post my own coverage soon, now that Denise (and others in the audience) are filing their own stories? Why shouldn't I?

Looks like it's going to get increasingly difficult to keep things off the record as time goes by.

I thought my battery backup for the Linux box was hosed - but it's not. After I had the power supply replaced awhile back, I plugged it into one of the power ports that isn't backed up by the battery. Sigh.....

I've had my issues with analysts before - now see what the Fishbowl has to say:

From this otherwise interesting C:Net article about the Sun v JBoss thing, comes the following choice quote:

Enforcing J2EE compliance is important, because IT buyers care about being able to move Java applications to different systems, said Ted Schadler, an analyst at Forrester Research. ..

False. Complete, and utter bullshit. The overwhelming majority of J2EE development is being done in bespoke systems, where the deployment platform is decided a long time before development even begins. Cross-deployment is never an issue. Cross-compatibility of developer skills is important, so you have a bigger pool of development talent to hire from, but developers are far easier to adapt to incompatibilities than software is.

Which leads me to note -

• It's easier to train Smalltalkers than Java developers
• Smalltalk is more productive
• Most people aren't using EJB anyway, for a variety of good reasons

So, exactly why would you use Java instead of Smalltalk? Oh yeah, because of the ease of finding all those highly skilled, fresh out of college OO experts. But there's an even better comment over at the Fishbowl on analysts:

Marcus J. Ranum comes up with a nice summary of Gartner on the firewall-wizards mailing-list. You've got to understand that most of the input into Gartner is from briefings arranged by the marketing departments of companies that are paying them to listen to their briefings. Basically, Garter sits at the apex of the hype food-chain; they consume pure hype and produce little s****-pellets of hype that is [sic] as dense as neutronium.

hehe. Well, you all know what I think.....

Charles Miller has some interesting thoughts on skins:

I consider skinnability to be a good reason to not use a program. Skinnability almost always means complete lack of standard controls, and useability that as been viciously compromised just so that some 13 year old boy can more easily graft on a bunch of stuff he scanned from an H.R. Giger calendar.

Go read the whole thing, especially the link above - it's a well thought out rant....

Yesterday, a storm interrupted my power - and I finally realized why the UPS wasn't keeping the Linux box up. Well, it turned out the storm did some damage after all. At first, when I noticed that the Linux box was off the net, I thought I had lost an ethernt card. But then I noticed that none of the wired devices were on the LAN either. After a bit of investigation, I discovered that the hub in the basement was the issue - removd that, and all was well again. Well, almost. Turns out one of my cables is bad, so one of the Replays is off the LAN. Sigh....

A lot of people have been commenting on MS' announcement that IE will soon (after the next service pack) stop shipping as a standalone product outside of the OS. This raises some interesting questions - does IE for the Mac go away then, or are they only talking about Windows? What about the AOL deal? Bitworking has some interestingt thoughts on why this might not be such a bright idea:

And let's not forget that upgrading the operating system costs money, versus downloading a new free browser off the web. If Microsoft thinks customers, given the choice of upgrading their operating system or changing to a different browser, are going to choose to pay money to upgrade their operating system, then someone needs to check what they're drinking in Redmond.

The impact also stretches to beyond Microsoft. If you are one of those banks that has an "IE only" web presence, you have to ask yourself how many people are going to upgrade their operating systems versus moving their money to a new bank. I know people that have switched banks for much less.

Yeah, my upgrade from ME to XP was hardly seamless. In fact, it wasn't so much an upgrade as a re-install from scratch...

Ongoing asks an interesting question - how do you know how many individuals are subscribed to your feed? First off, why do you care? Well, because subscribers are more committed to your content than the occasional browser - it's the same issue that people tracking websites have - people who come back time and again are more interesting to you (from a business perspective):

Well, OK, but I think that from a crass business point of view, someone who comes to ongoing because they're subscribed to my feed is worth more than someone who followed a link from anywhere out there, in exactly the same way that a subscriber is worth more to a magazine than a newsstand sale. So I continue to think that RSS is qualitatively distinguishable from the general ebb and flow of Web traffic, and if we could track it, that would be a good thing.

In this whole discussion, the fact that some users simply hate the idea of being tracked comes up. Ongoing cites an emailer:

I absolutely refuse to *store* any cookies on my computer that do not benefit me, but some web site owner or advertizer. I absolutely refuse to *use* cookies on any of my own sites - *unless* they're cookies that actually benefit the user (such as for storing configuration options for a bug tracker). And it would be nice if I sold enough books through my associationship with Amazon to cover the cost of hosting - but I don't. Even though I provide a lot of extra information and short reviews. But NO ads, and NO cookies. NO RSS-user ids either. Not ever. And that's a promise.

This is a serious thing to work through - vendors on the web (just like anywhere else) need to advertise their wares in order to sell, while at the same time, they need to minimize the amount of irritation said advertising causes end users. For instance, the recent trend towards what I call "page turds" - those on page popup ads. The ones that offer me a chance to close them - I don't mind those. The ones that don't - I hate them with a passion.

This isn't a new problem though - TV and Radio advertisers have been trying to deal with this for decades.

A reader sent me an interesting link trail:

Start at GotDotNet, Microsoft's .NET evangelism site. Select the Blogs link on the left. Select Don Box's blog. You may or may not know Don. He's an author of those really boring, yet thick, Microsoft technical books. I think he made a name for himself in the COM days.

Scroll down Don's blog to Microsofties and the XML Way.

Don references Doug Purdy's three slashdot pieces. Press on each of Part I, Part II, Part III.

Doug is using Squeak for XML processing. So there you have it. From Microsoft's .NET evangelism site right to Squeak. Clearly .NET is simply a false face to Smalltalk.

Heh.

After a reader request, I've made a few updates to the blog site:

• The Archives - By default, the archive page now shows the current month. There are navigation links for going back and forward - I'll be adding an interface to allow more specific navigation without url editing as well
• The Archives - layout has also been changed. I removed the poster name (since it's always me), and added a link to the category for the post
• Single item views - there's now a link to the next and previous items whenever you are viewing a single item

There's one more request, to allow the hiding of the blogroll and link lists. I'll get to that as well, but I have some other pressing work at the moment...

Then take a look at this:

The best fares are still going to be either booking in advance and staying a Saturday night or Air Canada have introduced new web fares that are for low cost domestic travel only and the fares are very low but are still on a first come, first serve basis.

If you have a minimum of 10 people travelling, we can register your meeting/convention with Air Canada and that will give you generous discounts on their regular fares (not weekend excursions or web fares) and would be applicable for travellers coming from outside of Canada. If your numbers are tight and some people venture ahead and book on the web before all is set up, their bookings will not count towards your minimum numbers.

Go check it out

Via Matt Croyden:

Shares of International Business Machines Corp. IBM.N were halted on Monday in over-the-counter trading, pending news.

IBM said in a statement shortly after the halt it is the subject of an investigation by the U.S. Securities and Exchange Commission. It said the probe is related to revenue recognition in 2000 and 2001.

I heard rumors about this last year; guess they had some weight....

Vassili Bykov pulled off running ST-80 version 2 - in a VW 7 image

Via Linux Today, Infoworld's Ed Foster tells us what the next tech suit will look like:

"Semi-Unabridged Webster Corp., publishers of the Semi-Unabridged Webster Dictionary of 1923, today announced it has granted a license of English language intellectual property rights to Microsoft Corporation. The licensing deal makes Microsoft the first company to ensure its products and services will be in copyright compliance with the use of English.

"'We are very grateful for Microsoft's acknowledgement of our intellectual property rights and for their support in our efforts to make English available to all for a reasonable licensing fee,' said Preston Gates Ellis, CEO and General Counsel for the recently-formed Semi-Unabridged Webster (SUW). Ellis acknowledged that Microsoft received a particularly favorable licensing due to its 'small investment' in his company, but declined to provide detailed numbers..."

Heh

I've posted a few times on the SCO vs. Linux circus, here, here, and here. Now PC Magazine's John Dvorak has weighed in with some panicky prose. He sees the possible outcomes as:

• VERSION A: SCO does own the Unix code. IBM took out a license for AIX. The court finds that IBM violated the license and IBM loses its case. SCO then asserts its intellectual property rights and goes after all Linux companies and users of Linux for fees. Confusion reigns. Linux dies.
• VERSION B: SCO pays no attention to code ownership and concentrates on contract violations with IBM. SCO wins, and uses the victory as a precedent to go after the tainted code used by others. Corporations panic and flock to Microsoft.
• VERSION C: IBM decides to stall the proceedings. It's using David Boies' old law firm. As the battle drags on, the Linux movement loses momentum, then dies.
• VERSION D: The case is thrown out of court because it has no merit, and SCO has no platform to stand on because it doesn't own the code. This is what the Linux community is hoping for. Good luck.

He's forgotten the most likely scenario and trumpeted up some really unlikely ones. The most likely outcome? IBM buys SCO and the whole thing goes away. Unless, of course, the rabbit hole goes deeper than we thought - in which case, SCO is just screwed, because an IBM payoff is their entire business strategy right now...

Smalltalk Solutions 2003 Would like to announce its technical conference schedule.

To view this year's complete technical conference schedule please go to the presentations site

Additionally, tutorials are available for $100USD each. To view the tutorial schedule, please go to the tutorials site

To register for the show, please visit our site at www.smalltalksolutions.com

Don't forget, early registration ends June 13th, 2003. See you all in Toronto!

For additional information please contact: Joy Murray

Via Simon St. Laurent we see how XML is not OO:

XML is pretty explicitly a rejection of an aspect of OO practice that Norm touches on only briefly: encapsulation. Everything accessible all the time is pretty clearly a hallmark of XML work. You can hide things if you want to, but it takes a lot more effort

I just posted updated schedule information on the show - and some people have asked about the fact that the site has the schedule as July 13-16, while all the technical schedules start on the 14th. The 13th (Sunday) is only for early registration (5 pm - 7pm) and exhibitor setup. The show itself is from July 14 - 16.

One of the content management tools is supporting RSS. Guess it's standard :)