Friday, July 18, 2003

The Parrot vs. The Python
Sterling Hughes points to a little challenge Guido Van Rossum (the creator of the Python programming language) and Dan Sugalski (one of the developers of the Parrot virtual machine for Perl, Python and possibly other dynamic languages) have going.  Dan claims Parrot will be able to run Python faster than the Python interpreter can by this time next year. Hmmm...
Permanent Link - No comments - Trackback URL

Automating Cumulus (AKA: PHP, VB, COM, Java, etc.)
I'm currently watching a script run through over 1,000 hi-res EPS images and convert those images to JPEGs in three different sizes onto a Windows share running on an AS/400 (aka iSeries). I thought it might be useful to jot down some notes on how I got to this finish line of sorts. [Read More]
Permanent Link - No comments - Trackback URL

Thursday, July 17, 2003

Solid email service
Since I tend to whine about tech vendors quite a bit here (especially lately), I thought it might be nice to point to at least one vendor I can't say anything but nice things about: FastMail. They specialize in email hosting for personal accounts; offering POP, IMAP and webmail access. Their support is great, the service is fast and reliable. The pricing is fair as well. Highly recommended!
Permanent Link - No comments - Trackback URL

Wednesday, July 16, 2003

Did I mention I HATE Visual Basic?
Hate is a strong word and I guess that is why it fits my feelings for Visual Basic. I've resorted to the lowest common denominator on the script I'm writing for our Digital Asset Management System (Canto's Cumulus 5 Enterprise) — Visual Basic 6. No, I don't want to mess with VB.NET. I'm using components that aren't .NET friendly and I'm not willing to play that game right now. So I'm using a language I always despise. It feels so foreign to me. I find what helps me when working with VB code is to think of how I would do something in another programming language and then do something completely different and bizarre in VB. I'm not just saying this because it's from Microsoft. I think C# looks pretty nice, but Visual Basic (even in .NET) just sucks. The more I use it, the more I find to hate about it.

As if that weren't bad enough, Canto has the most ridiculous restrictions on their API's. You can use the VB and AppleScript APIs free of charge (included with your license of the software), but if you want the Java API that will set you back at least $995 a year. What?!!! Yes, I let my reps at Canto know that this was one of the more backwards software related policies I've encountered lately. I actually asked them what year it was. I said if the year was 1985 I might have bought into their closed box mentality, but it's not and things have changed for the better. They gave me the excuse that the company is German owned and really paranoid about who gets access to their API's, etc. I thought about that after the phone call and came to the conclusion that it makes absolutely no sense. They're paranoid that someone might want to use and maybe even enhance the Cumulus product?! That's all you can really do with the API. And don't tell me someone wants to reverse engineer Cumulus. That application has changed much since the early-mid 90's. Canto isn't exactly taking over the world in the DAM market, so the weird obsession over keeping the API's locked up tight is puzzling. It's worth a laugh if nothing else.

Now it's back to finishing this lovely little script. Yeah baby!
Permanent Link - No comments - Trackback URL

Tuesday, July 15, 2003

Scoble wants feedback on the Windows platform
Scoble: Since a few Microsoft employees read my weblog, take a new tact. Instead of saying “stop locking us in the trunk” why don't you tell us what you'd like to see on the Windows platform?

I believe I did this for Scoble a while back.  I think that list is pretty legit. Or is he looking for something more specific to the Windows operating system only?
Permanent Link - No comments - Trackback URL

Monday, July 07, 2003

Jack of all trades and the custom CMS disease
There is a vendor I work with that really wants to get our CMS business. I've committed to CrownPeak Technology's ASP offering for at least one site. I enjoy working with this vendor who wants our business. They only put forth the utmost effort and never whine about the 100's of stupid questions I ask just about every week.

The problem I have with this vendor doing our CMS is the fact that we would be their first CMS customer. They want to use Merant's Collage CMS, which Merant presented a demo to me on. I was impressed with what they had but still had to come back to the fact that CrownPeak gives me a solid product with the experience to back it up. Plus, CrownPeak's core business is CMS. This other vendor is a jack of all trades. They don't have a single commercial CMS package installation under their belt. My concerns, as I've expressed them to this vendor, are clear.

Today some of the services this vendor provides have been a bit shaky. A server crashed causing one service to go down completely. Another service is behaving in odd manners. Yet another one is failing on a critical process and doesn't handle the errors properly. I can't help but think about the consequences of doing CMS with this vendor. If you aren't an expert at any one thing, then how can I depend on you to make a very important project (for us anyway) like CMS successful? I can't afford to have these types of hiccups. The users are suspect of IT these days. We don't communicate and we often don't appear to deliver on our promises or, worse yet, we don't deliver. I can't let this project fall into the category of “yet another failed IT project.” I need a company, a product and a service that delivers the goods. I need a company that has a history of delivering the goods. CMS isn't rocket science (as this vendor pointed out to me) but it can definitely turn into a mess real quick if you don't do it right.

I wanted to post this to remind myself that it's important to deliver on the work you already have with a client. Don't go chasing more business when the existing business you have with the client is beginning to crumble due to your not being able to deliver.

Also important to note here: If you are an agency of any sorts that provides web site development services, please do not continue to build custom CMS systems for every client you encounter. Buy, don't build. Standardize on one or two CMSes that will do the job for most of your clientele. Become experts on those systems and watch the smiles on your clients' faces as they get a solid CMS on time and for less money. You benefit and the client benefits. It's a win-win. This makes so much more sense than building custom CMSes that require extensive coding and testing and provide maybe 50% of what a commercial package/service can offer. If you're concerned your client will bypass you and go with the packaged solution, then you most likely don't provide much added value to your client. Sell your expertise. Sell the client on how much faster and how much cheaper you can implement a CMS for them with one of your preferred packages/services. Heck, sell the client on an ASP like offering. You host the CMS and provide everything the client needs to get up and running. The client doesn't sink a dime into infrastructure and you have a new reoccurring revenue stream. But please, whatever you do, don't sell the client on a custom built CMS. Enough said.
Permanent Link - 2 comments - Trackback URL

Feedster == Fast
Chris Sells points out that Feedster is a lot faster now. I tried it out and it screams. I'm going to start using it now.
Permanent Link - No comments - Trackback URL

Thursday, July 03, 2003

Gotta love PeopleSoft's quick thinking
ZDNet: PeopleSoft said over half those sales would come from contracts covered by a new customer-protection program that industry watchers said was a takeover defence in disguise.

I don't know if this is legal or not, but it's pure genius. You boost your sales and protect yourself from a hostile takeover all with the same tactic! I think there's a saying for this. Something about killing some birds with a heavy rock...
Permanent Link - No comments - Trackback URL

Dave Winer makes sense and extends an olive branch
I've stayed away from this whole Echo project that has a gang of the weblog community publically feuding with Dave Winer, who made much of what the Echo project is trying to reinvent happen in the first place. Dave has made mistakes over this as I'm sure we all have with things we're passionate about. But, he doesn't deserve the ridicule I've seen from a number of people. The attacks often get personal and don't address any real issues.

Dave wrote a short essay today on why he's concerned about the Echo project and what he's waiting to hear from those involved. I think he's being very resonable in this essay. He did, afterall, invent the very technologies that the people behind Echo seem hell bent to tear down and then bring back to much of its original form.
Permanent Link - No comments - Trackback URL

Wednesday, July 02, 2003

Strange article title
CNET: Study finds early Apache code so-so

The article is about a study a code quality analysis company did, Reasoning, on Apache 2 code.  It finds 0.53 defects per thousand lines of code compared to 0.51 for commercial software on average. So, doesn't that make both Open Source and commercial software so-so, not just Apache code or Open Source code as the article implies? It gets worse:

The comparable defect rate indicates that open-source software starts out as raw as proprietary software, but Reasoning said that ultimately open-source software has the potential to exceed proprietary software in quality.

I find it very interesting that a CNET editor decided to run the headline that Apache's code is “so-so”, while the article points out that, not only does Apache's early code have about the same defect rate as most commercial packages but it has potential to exceed proprietary software in code quality! This warrants an article title that implies Apache's code is not all that great? Hmmm...
Permanent Link - No comments - Trackback URL

Tuesday, July 01, 2003

A Java book I may actually buy
Magnus Mickelsson points to a book on utilizing Open Source J2EE technologies coming out in November. The book is written by active community members who know their stuff. I've been wondering how people put all these different technologies together in a way that works in the real world.
Permanent Link - No comments - Trackback URL

I have a better chance of buying The Pink Panther
My wife uses the cool computer at home — a Mac. It's an older iMac that currently runs OS 9.2 and it crashes at least a couple times a day thanks to Internet Explorer 5.x. I want to upgrade it to OS X but am waiting for the Panther version to come out. I'm not sure if it will run decently on this old G3 400mhz processor, but that's not my biggest problem at this point. My wife asked how much it would cost to upgrade and I said about $100 or so. She said, “I can live with the crashing.” Great, just great!
Permanent Link - No comments - Trackback URL

Monday, June 30, 2003

Asking the wrong question about Microsoft dropping stand alone IE

Ziv Caspi: IMHO, a decision to only add value to IE on future operating systems is simply making a good business move: there's no point in putting high-paid developers on a product that makes no income, right?

Ziv (and the numerous other posts/comments I've read along these lines) is correct. Microsoft sees no gain in putting highly paid developers on a product that makes no money. But, the most important question to ask following that insight is this: Then why did Microsoft devote so many resources to create and enhance Internet Explorer as it stands today? To answer this question one would have to begin to go down the path of Microsoft's legal woes, no?

Face it, Microsoft conquered an emerging market by leveraging a monopoly it already had in another market. They developed Internet Explorer and made it free for everyone.  They even went the extra mile for all of us and preinstalled it on Windows.  They took the lead in the browser market and currently have total domination (at least 90%.)  Now Microsoft has no incentive to pour resources into Internet Explorer.  It accomplished its goal of defeating the competition and can now concentrate on setting a new “standard” for users to adopt in the next version of Windows. That is why I find Microsoft's behavior with Internet Explorer so disturbing.

Permanent Link - No comments - Trackback URL

What happened to the Nucleus CMS site?
I've been trying (without any success) for the last week or so to reach the site for the excellent Nucleus CMS weblog tool that runs this site. I think the Nucleus project needs to move to Sourceforge. It's too good to die. Sigh.
Permanent Link - 2 comments - Trackback URL

PHP5 Beta 1 released

Sterling Hughes announces that PHP 5 Beta 1 is now available for download

Permanent Link - No comments - Trackback URL

The Colonel