a companion discussion area for blog.codinghorror.com

Has Joel Spolsky Jumped the Shark?


#144

Could you help me. I’ve been on a diet for two weeks and all I’ve lost is two weeks.
I am from Israel and too bad know English, give true I wrote the following sentence: Furniture plaza offers many clocks including our decorative wall clock, atomic wall clock, mantle clock, desk clock, table clock, and more clocks.

With respect :o, Rooney.


#145

Your site is very good. Thank you for the opportunity to sign your guest book.
I am from Emirates and learning to read in English, please tell me right I wrote the following sentence: Distinctive home decor, your home accessory resource! Find table clocks, wall clocks, and mantle clocks so your always just in time! We offer many decorative styles.

Thank you very much :P. Trent.


#146

I’m curious to know if he wishes that this article wasn’t posted given his involvement with working (cough, cough) being paid by Joel to run StackOverflow.


#147

So while I think VBS sucks, and writing your own web development language is a bad idea, I still think Joel knows a lot more than me about creating a successful business and that’s why I read his stuff.

That may be so, but during the whole CoPilot joke, he babbled incessantly about the greatness of New York, the wonders of interns writing code from a spec, and persistently refused to write a single sentence describing the business arguments that led them to decide it was a good idea, and for all the “written from scratch in 10 weeks” lies, noone wanted to discuss how much time and effort was spent writing the big-up-front-design spec before the interns arrived, or the time spent evaluating the open source software they used as a basis.

Joel may not be a moron, but he’s keeping a few secrets, and mostly on the business side of the topic. (That may be understandable, but licking his boots clean is a little excessive.)


#148

Reading all this A* blogger stuff on the .NET makes me really wonder who the coders are that actually acheive the great things?
The ones bitching about what isn’t good (Joel, Wasabi, ROR etc) or the ones who spend their bitching time on writing better code.


#149

I personally read Joel for entertainment, ideas, and that overall warm fuzzy feeling any programmer gets from someone who actually respects their profession and their work.

Virtually every site or blog I read will occasionally have content that I disagree with. Sometimes vehemently so. That’s healthy. If any community of people agrees on everything, it’s a sign of groupthink.

Joel is definitely an “old-schooler”. He’s wary of new toys. That’s healthy too. We need the conservatives as well as the early adopters. I think he prefers to wait until technologies are mature before he actually does business with them (as opposed to pet projects), which is perfectly reasonable to me. Ruby is not a mature product yet.

Perhaps he is having some internal argument and airing it out on his site; if so, it’s hardly a surprise given the number of programmers who simply will not listen to a bad word about Ruby. Perhaps he is subtly laying down the law for his company; if so, it’s his company, his perogative, and the post is an interesting glimpse into the politics of software companies.

Both his major products (FogBugz and CityDesk) are very good products and while they certainly aren’t perfect, they also sell at a tiny fraction of the price of products like Jira. Say what you will about things like Wasabi - I think it’s a bit ridiculous myself - but it’s enabled him and his company to provide great value at low cost. Isn’t that what every small/startup business wants?

Some of his posts may even border on the absurd (like the one about exceptions), but I think he’s entitled to a few “misses”, and no, I wouldn’t really say he’s jumped the shark or gone insane - he’s entitled to his opinions.


#150

" VB Sucks!! SUCKS!! "

This comment really sums up the dross quality of replies here.

He is talking about business decisions that are resulting in huge business profits for himself.

The contributors to this site are “techies” who will never understand the common need to program in a lower-demonitation language for ease of distribution.

THAT IS WHY YOU ARE STUCK IN A CUBICLE AND PEOPLE LIKE JOEL ARE RELEXING ON A TROPICAL ISLAND WITH A PINA COLADA!!


#151

Hmmmm, I hope Joel keeps blasting off at the
mouth because I can’t help but think someone
at 37 signals is going to get an idea -

“Hey, let’s write a bug tracking application.”"

Indeed, and no business will use it as they want their data on their own servers and don’t want to install a runtime, and your business fails, and you are flipping burgers AGAIN and life just seems so unfair and confusing and it you’d only think about the BUSINESS reasons and not the TECHNICAL reasons then maybe you wouldn’t be being bossed around by a kid 25 years younger than you.


#152

“That said, the whole Wasabi and dynamic language saga makes me feel like I just walked through a sewer. It’s not the kind of grand, intelligent softare lifecycle plan I’d expect out of a smart company like Fog Creek. That’s why it’s so jarring.”

That is because you are a TECHIE who does not understand that development decisions are made for BUSINESS REASONS. That is why you are unable to run your own software company. Do you see?


#153

This seems like the time the frog realizes he’s being boiled, and resigns himself to being the world’s best frog legs. I don’t believe anyone would look at the overall requirements today and say Wasabi is a good technical solution. Wasabi probably makes a lot more sense in view of the day to day craziness of selling a software product, but technical and business leadership need to balance today’s needs vs. the burden of design debt.

At some point, too, you need to fire the customers that are holding you back. To frame this a different way, let’s look at Windows and Mac. Windows built a large but unruly customer base by supporting everything and preserving backward compatibility. Mac OS’s come out regularly, but support a lot less hardware and haven’t been known for backwards compatibility. True, Microsoft made billions off of their pattern, but the huge weight of trying to please the crowd that wants their ten year old applications to work has contributed to the long delays in shipping Vista. Microsoft has finally become a little more pragmatic about dropping products and technologies (and the associated customers), but they still had to cut a lot of features they’d like to implement because they’re dealing with the legacy of decades of product deadlines.

Now, as a vendor, you need to make a decision on what kind of clients you want to sell to. As long I had enough volume to stay in business, I’d rather sell to Mac-like users than to cranky users who demand that I support ten year old drivers. I prefer developing with Microsoft technology, but if I were a vendor I’d rather sell to customers who don’t make me waste my time due to their poor decisions. There’s a range of workable business models there - in the OS field, and in the bug tracking field. There are plenty of people out there who continue to make a living building in VB, classic ASP, Access 97, etc. As time goes on, sane customers move on to newer technologies, and they’re stuck working for crazy people. It’s hard to write really good software for crazy people.

I’d be interested in seeing the numbers behind the business study that led to supporting classic ASP and PHP4 over ASP.NET and XSP on Mono. How many potential clients said they wouldn’t install .NET a few years ago, but actually have it now as part of a service pack? How many PHP shops might have considered installing Mono if FogBugz just wasn’t available for PHP4? Are there other products which will convert or run ASP.NET code under PHP, the way Mainsoft converts .NET applications to run under J2EE?

I disagree with Chris. I’ve looked at the FogBugz ASP code to customize our installation. If the Wasabi base code looks like this with some whizbang closures bolted on top, I pity the developers who have to work on this stuff. This application easy to hack (in the traditional sense, not the cracker sense), not customize.


#154

To balance what I’ve said - we use FogBugz for my work, and from an end user point of view, it is a fabulous product. It works really, really well. I wish they’s sell a reporting package instead of sticking to the “reports are bad” thing, but it is a really, really, really good product.

As long as their technical solution can support this level of product quality for years to come, Joel laughs last.


#155

i"Let’s put it this way. You have these constraints:

(1) Six years of code already written in VBScript

(2) Needs to run on customers’ Windows and Unix servers

(3) Minimize tech support costs

(4) Many customers refuse to install new runtimes on their servers, either because of IT policies or out of stubborness

What would be your solution?"
/i

Beatings for the people who allowed the VBScript situation to get six years old.


#156

@Chris, I think you’re misinterpreting my comments. In my comment, I talked about a continuum of technology support and the tradeoffs in customer base. I didn’t say “fire the customers who won’t install the latest beta”; .NET has been out for 4 years and is included with Windows 2003. It’s not bleeding edge stuff. You’ve misunderstood me if you think I’m advocating anyone firing half their customer base.

The time is approaching when customers will start to pass on FogBugz if it doesn’t run on ASP.NET. Wasabi may have been a good and necessary bridge solution, but as time passes it’s going to make as much sense as trying to sell a web product that runs on cgi.


#157

It sounded like madness when I first read the article, but as I’ve thought about it over the last 24 hours I think that Wasabi might be a very good idea. Functional languages are small (you can fit a scheme interpreter into less than 100k), easy to create and (if you can get people who know how to use them) powerful beyond the dreams of us mere .net developers if messrs Yegge, Graham et al are to believed. To add a new platform to the application they just need to write a new interpreter for the language which, if it’s well specced out, should be pretty easy to do. Your own language never becomes obselete as it can be reimplemented as the underlying platforms change without needing to change your application.

If you have very smart people working for you and you know you are always going to have very smart people working for you it’s not a crazy idea at all. It just looks like one at first glance.


#158

Keith, all you’ve done is show that developers such as you have zero business intelligence, and know nothing about running software companies for profit.

That is why your post was written from a cubicle and not an office with a view of Manhatten.


#159

Joel knows what he’s talking about. This is a bone-headed article with a lot of bone-headed follow-up.


#160

Wow! This is enlighting. Reminds me of certain cyber security people. #NCS18