We need an extreme like this to show the masses just how rediculous laws like this really
are. Until then, we're nothing but evil hackers and music pirates in their eyes.
I read somewhere once that publishers of quilting books are getting upset at grandmas who are sharing quilt patterns online. Many of the patterns are copyrighted.
Thanks for a thoughtful reply. I have just one more comment and a link.
First you say:
Lastly, source code is not documentation.
But well written code should be. The "short form" of Eiffel class, which shows the assertions is much better doc than any hand written manual. Java doc can be equally good.
I suppose we also need documentation that describes the overall structure of the system etc. But this can be kept at a fairly high level and it should not change much.
As far as hiring people, I've hired plenty and ran across few that interviewed well, but turned out to be a disaster.
You misunderstand. The book advocates a way to design websites so that it's easy to use and you don't have to waste your time trying to figure out how things work.
Imagine that the reply button on/. was at the very bottom of the page and you had to enter the number of the comment you were replying to.
Or think of doors, where you have to stop and figure out whether you need to push or pull to open it (sometimes instructions are taped to the door to make this task easier).
If you are interested in good user interfaces, I recomend this book.
This article appeared in NYT last weekend. The interesting thing it said that Sony makes about 4 billions per year on music sales, but about 40 billion on electronics sales (i.e. MP3 players, memory sticks, CD burners). How willing do you think would Sony be to reduce the income from consumer eletronics to satisfy their music division?
Obvious ways. Do not release it in digital form. Don't charge $20 for something that costs $0.50. Try not to rip off your customers. Let copyright expire after 5 years (not 120).
Have you read "The Future of Ideas" by Lessing? He has plenty of good ideas.
This brings up what I see as the main problem confronting web publishers and their audience
The real nice thing about the web is that everyone is a publisher and everyone is the audience. So the old model of the audience paying publishers doesn't work anymore.
Imagine that your hosting costs were reduced to zero. Would you still put up your web site? Would you if there were thousands of people who wanted to see it?
what exactly do you mean? The person can explain benfits of MI? Knows UML notation? What's "rigorous design"? Design that can be mathematically verified?
In other words, it produces a lot of cowboys. Don't expect your bazaar approach to be successful in the cathedral.
There are plenty of "cowboys" working on proprietary systems. You just don't know, because you cannot examine their code.
In OSS project the source code serves as the main communication medium between developers, it's the only documentation that never goes out of date.
Good software engineers are rare in any environment, but when you find them they can do amazing things.
As far as building "cathedrals" - I'd consider the Linux kernel and the Apache web server as pretty good examples of "cathedrals" built using the "bazaar" style of development. No?
What exactly does math teach you? It's not as much the nitty gritty calculus stuff that's important
as it is the skills you learn from doing math - being able to analyze problems, problem solving, and the most important skill a CS person
can have, thinking logically. It's skills like those that make you a good programmer.
It's much more than that. Math teaches how to handle abstractions. And software design is all abstraction - you try create solutions to real world problems from the abstract stuff of computing.
In my experience I found that programmers with lots of math background (way more than just calculus) make best software designers.
No will do. Like it or not, software is intellectual property. *My* intellectual property, the stuff that keeps food on the table and my kids
warm and dry. *I* write it, *I* decide how to license it. Open source is an option. One option.
Actually, if you are getting paid to write software most likely that software is not your IP. It belongs to the company that pays you. This is called work for hire.
Don't believe me? Try take the source with you to the next job.
The article also makes it seem like MS is advocating C# completely replacing C++, which it is not. C++ is still included in Visual Studio.NET and although MS is pushing C#, it's not going away in the MS toolbox.
Actually I thought about building a Slashdot like application on top of Freenet. That gives you distribution, security and mirroring, as well as sharing of bandwith etc. Now to find the time to actually do it...
What I see is that (and it has already started happening in the last year or so) all these little web sites will be bought up by a conglomerate and mergered together. The economics of this is quite smart. I mean, it's not really economical for one small company to have a 10K server and a 1k/month internet connection. If 10 of these sites have been merged together, they would come to 1/10 (maybe a little more) of the original cost
You assume that the sites must be centralized. I would not pay money to read Slashdot (especially since the stuff that interest me most is supplied free - i.e. the comments), but I'd happily contribute some bandwith and a server to help to run it.
Perhaps what we need is next generation USENET, where the postings and stories are distributed around lots of machines, without anyone needing to supply huge servers.
Think about it this way, would advertisers pay millions of dollars to advertise during the Super Bowl if they found out that there was a technology that a good population of TV watchers are using to block the super bowl ads?
That's why all these TV networks are having the "willies" about TiVO and similar devices.
I believe (and this is just a hunch) that we are seeing the end of "carpet-bombing" type of advertising and moving onto some new paradigm. Next time you see an expensive add on TV ask yourself, "Why am I being shown this add? What am I to buy?"
I've installed Linux on several old machines, with little trouble. However, none of these would run Gnome or KDE. Here is a list:
486 Toshiba laptop, 24M memory, 200M drive - installed Debian from floppies. X-windows with two bitplanes. I used it for Email and surfing the Web with Lynx.
P75 Toshiba laptop, 24M memory, 500M drive, CD, installed Debian "potato" - no X windows. I'm planning to turn this machine into a wireless router.
P120 no-name desktop, 48Meg, started with 800M drive. Red Hat 7.0. It's my home web, music etc server. No X-windows.
I guess having a low-end X-server and window manager would be nice. Wouldn't WindowMaker work?
Heh.....go ask your nearest music major to come to linear algebra class with you. See for yourself if musicians have better formal
reasoning than anyone else.
Actually a lot of mathematicians and scientist are quite musical. They are just better at science. I was a math major, with a music minor. I'd start the day in agebraic topology and end in 20th century music...
Finding that vital piece of information can be far more important than $20k, especially to a large organisation.
Very true. However, try convincing the average corporate bean counter. So, instead install "htDig" and actually show that you can make $20K, with a search engine on the intranet. Once the people who use and need it are "hooked", you can proceed to getting Google (after all you should have supported software for "mission critical" functions, and you are much too important to administer htDig:-))
Try htDig. It does all these things and is free software. I used it on a corporate intranet in the past. Not as good as Google, but you can't argue with the price.
I read somewhere once that publishers of quilting books are getting upset at grandmas who are sharing quilt patterns online. Many of the patterns are copyrighted.
Is this extreme enough? Let's arrest grandma!
When I interview people I'm very suspicious of 4.0 GPAs. Just means the person didn't take any difficult courses. :-)
Lastly, source code is not documentation.
But well written code should be. The "short form" of Eiffel class, which shows the assertions is much better doc than any hand written manual. Java doc can be equally good.
I suppose we also need documentation that describes the overall structure of the system etc. But this can be kept at a fairly high level and it should not change much.
As far as hiring people, I've hired plenty and ran across few that interviewed well, but turned out to be a disaster.
Finally, regaring hiring I found this article The Guerrilla Guide to Interviewing really helpful.
You misunderstand. The book advocates a way to design websites so that it's easy to use and you don't have to waste your time trying to figure out how things work.
Imagine that the reply button on /. was at the very bottom of the page and you had to enter the number of the comment you were replying to.
Or think of doors, where you have to stop and figure out whether you need to push or pull to open it (sometimes instructions are taped to the door to make this task easier).
If you are interested in good user interfaces, I recomend this book.
Have you read "The Future of Ideas" by Lessing? He has plenty of good ideas.
The real nice thing about the web is that everyone is a publisher and everyone is the audience. So the old model of the audience paying publishers doesn't work anymore.
Imagine that your hosting costs were reduced to zero. Would you still put up your web site? Would you if there were thousands of people who wanted to see it?
* strong OO principles
* rigorous design
* excellent documentation
what exactly do you mean? The person can explain benfits of MI? Knows UML notation? What's "rigorous design"? Design that can be mathematically verified?
In other words, it produces a lot of cowboys. Don't expect your bazaar approach to be successful in the cathedral.
There are plenty of "cowboys" working on proprietary systems. You just don't know, because you cannot examine their code.
In OSS project the source code serves as the main communication medium between developers, it's the only documentation that never goes out of date.
Good software engineers are rare in any environment, but when you find them they can do amazing things.
As far as building "cathedrals" - I'd consider the Linux kernel and the Apache web server as pretty good examples of "cathedrals" built using the "bazaar" style of development. No?
It's much more than that. Math teaches how to handle abstractions. And software design is all abstraction - you try create solutions to real world problems from the abstract stuff of computing.
In my experience I found that programmers with lots of math background (way more than just calculus) make best software designers.
Actually, if you are getting paid to write software most likely that software is not your IP. It belongs to the company that pays you. This is called work for hire.
Don't believe me? Try take the source with you to the next job.
Does C++.NET still support multiple inheritance?
Right. BUt what if the actual act of recording will be possible only if you bypass access controls - otherwise DVRs will not work.
You're right. It's harder you have to know how to type...:-)
Maybe because we are not just readers. We are also the writers and the editors (watch out, I got moderator points and I know how to use them!)
You assume that the sites must be centralized. I would not pay money to read Slashdot (especially since the stuff that interest me most is supplied free - i.e. the comments), but I'd happily contribute some bandwith and a server to help to run it.
Perhaps what we need is next generation USENET, where the postings and stories are distributed around lots of machines, without anyone needing to supply huge servers.
Think about it this way, would advertisers pay millions of dollars to advertise during the Super Bowl if they found out that there was a technology that a good population of TV watchers are using to block the super bowl ads?
That's why all these TV networks are having the "willies" about TiVO and similar devices.
I believe (and this is just a hunch) that we are seeing the end of "carpet-bombing" type of advertising and moving onto some new paradigm. Next time you see an expensive add on TV ask yourself, "Why am I being shown this add? What am I to buy?"
The only environment I'm addicted too is Emacs...
486 Toshiba laptop, 24M memory, 200M drive - installed Debian from floppies. X-windows with two bitplanes. I used it for Email and surfing the Web with Lynx.
P75 Toshiba laptop, 24M memory, 500M drive, CD, installed Debian "potato" - no X windows. I'm planning to turn this machine into a wireless router.
P120 no-name desktop, 48Meg, started with 800M drive. Red Hat 7.0. It's my home web, music etc server. No X-windows.
I guess having a low-end X-server and window manager would be nice. Wouldn't WindowMaker work?
Actually a lot of mathematicians and scientist are quite musical. They are just better at science. I was a math major, with a music minor. I'd start the day in agebraic topology and end in 20th century music...
Very true. However, try convincing the average corporate bean counter. So, instead install "htDig" and actually show that you can make $20K, with a search engine on the intranet. Once the people who use and need it are "hooked", you can proceed to getting Google (after all you should have supported software for "mission critical" functions, and you are much too important to administer htDig :-))