Joel Spolsky's talk at Yale, part 1(joelonsoftware.com)
joelonsoftware.com
Joel Spolsky's talk at Yale, part 1
http://www.joelonsoftware.com/items/2007/12/03.html
5 comments
I have read many of Joel's posts and most of them I find quite lame. There is no substance. His world seems to be centered around how to developing boring software optimally rather than making next generation of software which nobody has ever made. This is apparent from his ideas against the grad school where you actually learn to do new things and also looking at the kind of software which fog creek makes which seriously makes me want to yawn. Why is he given so much importance by the people? I really don't understand.
Yeah it looks so much sexier to throw together some lines of rails code over the weekend and become a millionaire over night. However, most of the time getting successful actually entails some hard work. So don't laugh about Joel's posts too soon.
Besides, his latest feature for estimates sounded quite interesting (when he described it, anyway). Certainly more interesting than yet another "vote me up, vote me down" Web 2.0 application.
Besides, his latest feature for estimates sounded quite interesting (when he described it, anyway). Certainly more interesting than yet another "vote me up, vote me down" Web 2.0 application.
Web 2.0 sucks, I agree. Its suckage does not preclude bug tracker suckage, though. The GP had a point, which is that boring software is boring, and really, it would be better to read stuff about interesting software, by people who regularly write interesting software. There is interesting software floating around; the problem is, most of the people who write it don't have time to blog about it. They sometimes do write about it, but usually in that case it comes in the form of either patents or academic papers, both of which are unnecessarily hard to read.
Still, I enjoy Joel's articles. And some stuff he built was neat, like "how to ship things". Reminds me that I still haven't watched the video.
I also don't think that Web 2.0 sucks, but it sucks if people look down on real software. two-week rails projects will only get mankind so far, I suspect.
I also don't think that Web 2.0 sucks, but it sucks if people look down on real software. two-week rails projects will only get mankind so far, I suspect.
Web 2.0 sucks in the sense that if you could be working on real things, it sucks to be working on two-week rails projects. Clearly, sometimes you have to do two-week rails projects or other equivalently boring things in order to get rich, but that doesn't make them any more interesting to talk about, from a hacker's perspective.
Joel has lot's of money, you don't (I presume - wildly)
Software that changes the world is a lottery with slightly worse odds. Software that solves boring problems effectively and elegantly is a good way to make a living.
I'd put Fog Creek up against the latest web 2.0 fad company lasting until 2020 any day of the week. There is some merit behind that.
Software that changes the world is a lottery with slightly worse odds. Software that solves boring problems effectively and elegantly is a good way to make a living.
I'd put Fog Creek up against the latest web 2.0 fad company lasting until 2020 any day of the week. There is some merit behind that.
As the saying goes where there's muck there's brass. I believe there is money to be made in niche markets where there is a need for software that solves boring problems effectively. The way I see things are going we're too tied up building the next "poke your friend in Facebook" app that we tend to forget that there are real problems being experienced by people/businesses that are begging for a solution. Not a knock against face book developers but it would be nice to use that kind of brain power to solve "mundane" problems.
Wait first wasn't there a PG essay about this already? Shame on you!
Wait first wasn't there a PG essay about this already? Shame on you!
true. their copilot app (built by interns largely) is actually pretty cool.
and the spec he put together which outlines copilot is a pretty good read:
Spec: http://www.joelonsoftware.com/RandomStuff/copilot_spec.pdf
Article: http://www.joelonsoftware.com/articles/AardvarkSpec.html
Spec: http://www.joelonsoftware.com/RandomStuff/copilot_spec.pdf
Article: http://www.joelonsoftware.com/articles/AardvarkSpec.html
yeah, actually that whole thing was fascinating to watch. The process actually worked. I think there is a bit much weight put on the process though, lets face it, the people involved were brilliant. They could have started with a wet towel and got something great out of it.
That reminds me, there was a DVD - anyone watch that? I have been meaning to buy it.
That reminds me, there was a DVD - anyone watch that? I have been meaning to buy it.
rather than making next generation of software which nobody has ever made.
Yeah, good luck. I think Carmack performed the last instance of that right around 1993. Most Web 2.0 applications that I've seen are basically CRUD apps.
Yeah, good luck. I think Carmack performed the last instance of that right around 1993. Most Web 2.0 applications that I've seen are basically CRUD apps.
How about DabbleDB?
they have a bit of competition now
http://www.oncetechnologies.com/radix/
http://www.aphplix.org/
probably 100's of others (probably not as good though). Dabble db is pre web2.0 though.
probably 100's of others (probably not as good though). Dabble db is pre web2.0 though.
ughh
there is a difference between making money and making money doing cool things. Joel makes moeney by doing really boring things really well. Web2.0 stuff makes me want to yawn.
The interesting future lies more in the fields of Machine Learning, Computer Vision, 3D graphics, Bioinformatics, and what not. Things which are not trivial to do and for which now slowly and steadily we have the computing power to make it happen on the standard desktops.
trying to make money using web2.0 type stuff is like lottery i think. There are so many people trying to do it.
The interesting future lies more in the fields of Machine Learning, Computer Vision, 3D graphics, Bioinformatics, and what not. Things which are not trivial to do and for which now slowly and steadily we have the computing power to make it happen on the standard desktops.
trying to make money using web2.0 type stuff is like lottery i think. There are so many people trying to do it.
Splosky's elitism would be easier to take if his company weren't so pedestrian.
It seems that, if anything, he's attacking the ivory tower mentality that a lot of geeks have. He's looking at the whole software picture, rather than just the utilitarian aspect.
Spolsky has a lot of good things to say, but he does take it a little too far sometimes. His technical credibility took a major hit in my eyes when I heard about the mess that is Wasabi. Still, the way he runs his company is very innovative, and he has a lot of good things to say about enhancing productivity. Overall, I wouldn't say he's a great technical resource, but in terms of running a successful niche business, he's got some good ideas.
I think Wasabi is very misunderstood. Creating a meta language that "compiles" down to PHP or ASP makes a lot of sense. Don't we need every advantage we can get? If doing that speeds up his development cycle and helps him support both Linux and Windows customers, I wouldn't call that a mess.
It's unorthodox maybe, but not a mess.
It's unorthodox maybe, but not a mess.
what the hell is wasabi? VBScript with picture functions?
It's his secret weapon to build his product both in ASP and PHP. I bet if you were him you'd say "go away customer, I only write my software in Rails/LISP/Haskell".
Jane says that her car's engine has a problem.
Jane works at a bakery.
Therefore, Jane is wrong.
Jane works at a bakery.
Therefore, Jane is wrong.
Real programmers work on bug trackers. You can move the columns in tables around, for gods' sake! Why don't you people appreciate genius when you see it?
I bet it's because you don't understand pointer arithmetic. Morons...
I bet it's because you don't understand pointer arithmetic. Morons...
And yet, I spent 5 years working for a company using Peregrine ServiceCenter, the most god-awful piece of junk software I've ever seen, for trouble-tickets. I actually once measured how much time was lost in a day because of its suck. It wasn't insignificant, and it didn't even begin to cover the emotional effects on productivity. I hated it with a passion, to the point where it would be personal revenge for me to drive Peregrine out of business (at least out of the trouble-ticket software business).
I feel the same, only my hatred is saved for lotus notes.
I have been forced to use terrible software before so I can empathize.
This is why $DEITY invented Trac.
This is why $DEITY invented Trac.
And recursion. You ignorant clod. You forgot recursion. Clearly you will never rise to the level of writing bug trackers in VBScript.
Not straight up VBScript -- a poorly designed Basic dialect that is less powerful than VB.NET and far less general.
Actually I didn't really mind the wasabi thing. Odd, yes, but hey, they already had it all working in asp.
One thing from the article I found interesting is that a very detailed spec is equivalent to the program. That is, I found it interesting when I "discovered" it a few years ago.
Let's try a little poll: if you found that interesting and already knew it, upvote this comment. If you found it interesting but didn't know it, downvote it. If you didn't find that interesting, don't vote.
Let's try a little poll: if you found that interesting and already knew it, upvote this comment. If you found it interesting but didn't know it, downvote it. If you didn't find that interesting, don't vote.
"The old testers at Microsoft checked lots of things: they checked if fonts were consistent and legible, they checked that the location of controls on dialog boxes was reasonable and neatly aligned, they checked whether the screen flickered when you did things, they looked at how the UI flowed"
That's a crack-up. Please, regale us with more tales of the incredible quality of Microsoft products of the 90's.
That's a crack-up. Please, regale us with more tales of the incredible quality of Microsoft products of the 90's.
I think deserve some credit for getting a hold on the hardware diversity mess. A problem that Apple never had to solve.
They didn't customize their applications for different types of hardware, and drivers are made by the manufacturer, so this is a non-point.
Windows et al. were ugly, buggy and crashing due to software problems, not hardware.
Windows et al. were ugly, buggy and crashing due to software problems, not hardware.
They are used by 90% of the marketplace afterall.
they looked at how the UI flowed
UI may flow well, except it may flow in a wrong direction.
In 90's there were pieces of software that were close to the definition of "crappy" in terms of bugs, UI flow and even screen flickering, yet they delivered such a value that users forgave their mistakes. First that comes to mind is Macromedia Dreamweaver, also Cakewalk Pro Audio.
UI may flow well, except it may flow in a wrong direction.
In 90's there were pieces of software that were close to the definition of "crappy" in terms of bugs, UI flow and even screen flickering, yet they delivered such a value that users forgave their mistakes. First that comes to mind is Macromedia Dreamweaver, also Cakewalk Pro Audio.