I was quite impressed by the fact that Apache can cram all the functionality into ~59k lines. So besides defect rate, I would like to know how many lines of code the commercial package had... 0.51 defects per 1000 lines sounds good, unless there are 1,000,000 lines more code in the commercial package.
>..the ability to deliver software capabilities
> encapsulated within markup language (HTML/XML)
> communications units delivered over a logical
> Internet connection to a browser which then
> executes scripts to automatically perform one
> or more tasks on a computer.
Is it me, or does this sound like an excellent means of transferring viruses TO a computer? Perhaps a virus-maker could claim prior art?
Sorry for the original spin on the story, but I had never come across this type of open-core development before. Another link someone posted is OpenCores.org which also has an open PIC core as well as some DSP cores, and many others.
There are a lot of companies that don't really care about chips and OSes (like those developing automotive entertainment systems), they just want to push a product out the door quickly and cheaply. My original point was more that an open-* approach (if more well known) could help lower the entry costs for new companies - potentially stimulating innovation as has happened within the Linux community.
You bring up a good point - one that is also found in patent law I believe. If you choose to patent a device, you must make blueprints, drawings, etc. available to the public... there must be some kind of disclosure so that the public can benefit from the monopoly position it is offerring to your now-patented device.
The other option is to keep your new invention a "trade secret." The stereotypical example of this is the machines that print the M's on M&Ms - they never patented it so that their competitors never got to see how they did it (and thus were unable to copy it). But M&M Mars has to take extreme efforts to protect this themselves - employees must sign non-disclosures and other legal documents to insure that they don't leak the "trade secret". Since you haven't offered any benefit to society (your blueprints are behind locked doors), then society doesn't offer any protection back to you.
Perhaps there could be a choice (1) copyright a book the usual way and allow fair use to licensees but also gain some simple protections from the government, or (2) protect your own book using your own methods including any access controls you like but have no DMCA to protect you.
It would seem to me that anyone can print a paper copy of web-based books/documents on their printer. Granted that toner and paper costs add up, but that's the case for a print-shop too. So your only profit margin would really be the difference between what your (large) print-shop and the home-equivalent -- things like volume discounts on raw materials. I suppose there's the cost of the binding, but if you're looking at documentation, people might not want $25 binding fees - a 3 ring binder at $1.50 might be all I need.
Having said that, there *is* some real value in this type of service - I just wonder about the costs and what people would be willing to pay.
Another thing to worry about is Kinkos/Staples/ etc. If vanity presses and print-on-delivery gets to be profittable, a lot of already established companies can jump right in. And many people would prefer to drop their file off at the local copy center than ftp it to someone they don't know.
I'd have to agree - building a good set of course materials is very hard, and almost no fun. Since you know the material so well, you will probably leave a ton of info out - stuff only a second person (ie. editor) might notice.
You should also consider what it might cost to print out 30-50 pages of open source documentation (per student) on your office printer plus a binder of some form to keep it together - then paying for a book doesn't look so bad anymore.
I've got to believe some of the $30 Dummies/Idiots/Nutshell books are decent enough to teach an intro-level course around.
Re:consider giving your time, too.
on
Geek Charities?
·
· Score: 1
as usual: IANAL... and I'm not trying to suggest that a tax break is the only reason to donate to a charity, but...
From what I recall, donations of time are also tax deductible just like cash donations. I don't know how one figures out the hourly rate, but it's certainly possible to do. Then you give the organization the benefit of your time/work/elbow grease and you still get a small tax break.
Beta got clobbered by VHS (even though most people agreed it was a superior technology)
a friend of mine once bought a laser disk video system 12 years ago (with the 12" disks, remember those? back when "laser" was still a catchy buzzword)
DIVX anyone?
I've owned 8" disk drives
and 5.25" disk drives
and friends of mine had to use cassette tape to store their programs (good ol' C-64!)
for a more recent example, look at Iomega - their Zip and Click drives - while I love the company, if it should ever go bankrupt, or decide that a given drive is not worth pursuing anymore, you're outta luck and your data could be irretrievably lost
All of these media formats went the way of the dodo - although the data/music/images stored on them are still very valuable.
I personally have re-purchased several music albums twice and possibly even three times - simply to have use of it in different formats. My friend with the laser disk video system paid over $100 for each movie - and while the hardware system still works, when it breaks that investment in the media (or licensing of the media) is gone.
How can anyone not understand how this will impact their lives? There is an enormous amount of historical evidence to suggest that you will have to re-purchase items that you already have a legal license for... unless you have the right to make back-up copies.
And if the hardware company you're reliant upon suddenly goes belly up, you better have a way to reverse engineer your back-up system.
The music industry should have seen this problem a long time ago - CDs and DATs, and hence the possibility of digital reproduction, has been around for 12-15-20? years. But when did the RIAA decide to look at secure digital music? A year ago, maybe two?
If I were an artist, I'd be a little ticked off that they hadn't protected my songs better. They had the money, they should have seen it was coming, but they didn't resolve the issue. Now MP3 and Napster (and Gnutella) are here, and it's gonna be hard/impossible to get rid of them.
I agree that Napster aids kids in ripping of copyrighted works, but the labels haven't made it easy (or at least not reasonable) to purchase songs over the net. The labels have not upheld their financial duties to their artists.
I recall reading a similar election-year article in Discover (several years ago) about how the US system was GOOD in that it did not require simple majority. In order to win the election, you have to be reasonably supported by many states, not SIMPLY 51% of all people.
The authors likened this to winning the World Series - it's not enough to score the most runs, you have to score those runs in such a way that you win 4 games. In fact, if you lose a game by 20 runs, then win the next 4 by 1 run each - you've "lost" in one sense, but most would agree who's the stronger team.
A lot of "systems research" is now going on in metacomputing (see the interview with Greg Lindahl, or the Legion Project at U.Va, http://legion.virginia.edu, or Globus, http://www.globus.org). These groups are attempting to develop wide-area distributed computing systems but are purposely avoiding the low level systems issues since they cannot guarantee them to be consistent across Solaris/IRIX/AIX/Linux/etc. This is real systems research, if you want to borrow someone else's CPU power, potentially someone whom you've never met or spoken with, you need a very flexible system.
Before you can comment on a lack of "systems" research, you have to define what a system is. Most of us would probably agree that there is little innovation in low-level, nitty-gritty device driver systems research. But if you bought a new machine and all it came with were device drivers, would you consider that a complete system?
Slashdot Mirror
I was quite impressed by the fact that Apache can cram all the functionality into ~59k lines. So besides defect rate, I would like to know how many lines of code the commercial package had ... 0.51 defects per 1000 lines sounds good, unless there are 1,000,000 lines more code in the commercial package.
> ..the ability to deliver software capabilities
> encapsulated within markup language (HTML/XML)
> communications units delivered over a logical
> Internet connection to a browser which then
> executes scripts to automatically perform one
> or more tasks on a computer.
Is it me, or does this sound like an excellent means of transferring viruses TO a computer? Perhaps a virus-maker could claim prior art?
There are a lot of companies that don't really care about chips and OSes (like those developing automotive entertainment systems), they just want to push a product out the door quickly and cheaply. My original point was more that an open-* approach (if more well known) could help lower the entry costs for new companies - potentially stimulating innovation as has happened within the Linux community.
The other option is to keep your new invention a "trade secret." The stereotypical example of this is the machines that print the M's on M&Ms - they never patented it so that their competitors never got to see how they did it (and thus were unable to copy it). But M&M Mars has to take extreme efforts to protect this themselves - employees must sign non-disclosures and other legal documents to insure that they don't leak the "trade secret". Since you haven't offered any benefit to society (your blueprints are behind locked doors), then society doesn't offer any protection back to you.
Perhaps there could be a choice (1) copyright a book the usual way and allow fair use to licensees but also gain some simple protections from the government, or (2) protect your own book using your own methods including any access controls you like but have no DMCA to protect you.
Having said that, there *is* some real value in this type of service - I just wonder about the costs and what people would be willing to pay.
Another thing to worry about is Kinkos/Staples/ etc. If vanity presses and print-on-delivery gets to be profittable, a lot of already established companies can jump right in. And many people would prefer to drop their file off at the local copy center than ftp it to someone they don't know.
I'd have to agree - building a good set of course materials is very hard, and almost no fun. Since you know the material so well, you will probably leave a ton of info out - stuff only a second person (ie. editor) might notice.
You should also consider what it might cost to print out 30-50 pages of open source documentation (per student) on your office printer plus a binder of some form to keep it together - then paying for a book doesn't look so bad anymore.
I've got to believe some of the $30 Dummies/Idiots/Nutshell books are decent enough to teach an intro-level course around.
From what I recall, donations of time are also tax deductible just like cash donations. I don't know how one figures out the hourly rate, but it's certainly possible to do. Then you give the organization the benefit of your time/work/elbow grease and you still get a small tax break.
- 8 tracks came and went
- vinyl LPs came and went
- cassettes are rapidly giving way to CDs
- Beta got clobbered by VHS (even though most people agreed it was a superior technology)
- a friend of mine once bought a laser disk video system 12 years ago (with the 12" disks, remember those? back when "laser" was still a catchy buzzword)
- DIVX anyone?
- I've owned 8" disk drives
- and 5.25" disk drives
- and friends of mine had to use cassette tape to store their programs (good ol' C-64!)
- for a more recent example, look at Iomega - their Zip and Click drives - while I love the company, if it should ever go bankrupt, or decide that a given drive is not worth pursuing anymore, you're outta luck and your data could be irretrievably lost
All of these media formats went the way of the dodo - although the data/music/images stored on them are still very valuable.I personally have re-purchased several music albums twice and possibly even three times - simply to have use of it in different formats. My friend with the laser disk video system paid over $100 for each movie - and while the hardware system still works, when it breaks that investment in the media (or licensing of the media) is gone.
How can anyone not understand how this will impact their lives? There is an enormous amount of historical evidence to suggest that you will have to re-purchase items that you already have a legal license for... unless you have the right to make back-up copies.
And if the hardware company you're reliant upon suddenly goes belly up, you better have a way to reverse engineer your back-up system.
If I were an artist, I'd be a little ticked off that they hadn't protected my songs better. They had the money, they should have seen it was coming, but they didn't resolve the issue. Now MP3 and Napster (and Gnutella) are here, and it's gonna be hard/impossible to get rid of them.
I agree that Napster aids kids in ripping of copyrighted works, but the labels haven't made it easy (or at least not reasonable) to purchase songs over the net. The labels have not upheld their financial duties to their artists.
I recall reading a similar election-year article in Discover (several years ago) about how the US system was GOOD in that it did not require simple majority. In order to win the election, you have to be reasonably supported by many states, not SIMPLY 51% of all people. The authors likened this to winning the World Series - it's not enough to score the most runs, you have to score those runs in such a way that you win 4 games. In fact, if you lose a game by 20 runs, then win the next 4 by 1 run each - you've "lost" in one sense, but most would agree who's the stronger team.
Before you can comment on a lack of "systems" research, you have to define what a system is. Most of us would probably agree that there is little innovation in low-level, nitty-gritty device driver systems research. But if you bought a new machine and all it came with were device drivers, would you consider that a complete system?