ICANN is the root problem here, and in many other issues. Specifically, ICANNs complete lack oversight over registrars. This isn't s registrar problem, it is a web hosting problem (or a DNS service problem in the case of subdomains) that happens to involve a hosting company that is also a registrar. Planting ads on 404 pages could be done by any (scummy) hosting company. Registrars that don't provide hosting can't monkey with 404 pages. The problem doesn't involve the registration of the domain name, it's the optional services (hosting/DNS) provided after registration where the problem arises, so I don't see how ICANN has anything to do with it. I'm not saying ICANN doesn't have shortcomings, I'm just saying this isn't one of them.
she does use some crazy word functionality for tracking edits. Different parts of a document are highlighted according to when and by whom they were eddied by You mean like Edit / Changes / Record?
The "percentage of website users that are running Linux" is meaningless too, as many or maybe even most are having Firefox or Konqueror report to the server that it's IE on Windows, jusr for the websites written by idiots that redirect to a page that says "you must use IE", or are using WINE.
It's foolish to have your browser report that you're running Linux, and few GNU users are fools.
As one of the "foolish" Linux people who doesn't configure his browser to lie about running on Linux, I can say that I've encountered rather few websites worth going to that require such nonsense. If you are doing the majority of your browsing with IE on WINE, or you have your Firefox/Konqueror configured to tell every website you visit that you are running IE on Windows (rather than just the sites that really require it, which probably won't work anyway because they probably use Active-X), then you are just perpetuating the idea that it is OK to create IE/Windows-only websites.
Still meaningless. It doesn't count the half dozen or so copies I gave away.
That depends -- do the people using those half dozen copies surf the web? My point is that the 0.8% number is very similar to what I am seeing as "percentage of website users that are running Linux." If that is the measurement they are talking about, how the people got their copies has nothing to do with it. They are measuring how many people are using Linux to surf the web, not how many obtained it in any particular way.
I don't know where they get their stats from, but if I check the webserver logs for magportal.com (a fairly generic audience), I see 0.77% of unique IP addresses having "Linux" in the user-agent string over the past week.
I'm pretty sure regular contact lenses don't move when you blink, why would these?
Yes, regular contacts do move when you blink. It is most noticeable when your eyes are dry (e.g. when you are tired) because the lenses don't slide back into place as easily. Of course, if you have a weak prescription you probably won't notice the blurring when you blink because your lenses weren't doing all that much in the first place, so it doesn't matter if they are a little off. I'm not an optometrist (IANAO?), I just wear really strong contacts.
Aside from all of the other problems people have pointed out, what happens when you blink? The display moves and then settles back into position? Movement of the lens isn't a big deal when the whole thing is clear, but I would imagine it would be really annoying when there is a display on it.
Why? My sister became a freelance last year. I set her up a Debian Box with postgesql and SQL Ledger. I did the technical side. Both my dad and my sister have quite extensive accounting experience and they set the business side up in no time.
That's great if you're an accountant. If you use an external accountant, they'll want you to provide data in a format they can work with, which probably means Quickbooks for a lot of people.
I've seen claims that the ext3 filesystem updates its journal every 5 seconds. If you are running ext3 on the drives, maybe you're not losing the connection because ext3 wakes it up too often for power save to kick in.
However, it doesn't matter if the company doesn't modify the source. The point of the AGPL is that if a user on the intranet uses this web service, then he or she is supposed to have access to the source. So even though company Y doesn't make any modifications, the source is still available, and in turn can distribute the source by the terms of the AGPL. I haven't read the whole AGPL license, but section 13 says:
Notwithstanding any other provision of this License, if you modify the Program, your modified version must prominently offer all users interacting with it remotely through a computer network (if your version supports such interaction) an opportunity to receive the Corresponding Source of your version by providing access to the Corresponding Source from a network server at no charge... with emphasis added by me. So it seems that it does matter whether or not the company modifies the source, since it says "if you modify" not "if you use a modified version." Company X must offer source to its employees because it modified the code, but company Y can use the very same code without offering source to its employees if it makes no modifications.
In fact, this suggests a loophole -- company X could form a separate company to create the modified version of the software, that separate company could give a copy of the modified software to company X, and company X could now use it on its intranet without offering source to its employees.
But yes, that employee may redistribute the source under the terms of the AGPL. But, even if that employee leaves and works for a competitor, that competitor will need to abide by the same terms. Except that the competitor wouldn't be required to distribute the source unless the competitor made its own modifications. Suppose company X takes some generic AGPL code, adds a lot of industry-specific code to it (not of any interest to the original project, but very useful to companies in its industry), and uses it on its intranet. Any employee of X has the right, but no obligation, to obtain the source and distribute it, so the employee could sell it (i.e. "I could give it to you, but I won't unless you offer me a job with a $500k signing bonus") to another company in the same industry, right? If company Y obtains the code from an employee of X, company Y doesn't have to distribute the source unless it changes it, and it may not need to make changes if company X did a good job. So, employees of company X become special assets, because they have rights to distribute the code that other users of the exact same code (i.e. employees of company Y) don't have.
IANAL, but it seems to me you only have to give the "users" the source. On an intranet all the users are in the company, so there is no need to release the source outside the company. But, do you need to release it to any employee who asks for it, and may they then take it with them when they leave the company and go to work for a competitor?
They probably incorrectly labeled all Linux Firefox users as Windows Users... Firefox has a global 25% share (probably in excess of 90% linux browser share)
For the Linux count to be off by a factor of 100 due to not counting Firefox, Firefox would need 99% market share among Linux users. A quick and dirty analysis of a very small sample of the logs from MagPortal.com gives the breakdown (unique IP addresses with "linux" in the user-agent string [case insensitive]): FireFox: 53% Gecko (FireFox + Seamonkey + Minefield + etc.): 72% Konqueror: 11% Opera: 2%
Those numbers could easily be off by a fair amount, and I haven't excluded things like Wget or Java from the total count when computing percentages, but I would be surprised if Konqueror would come out below 1% of Linux users on a better sample.
Microsoft will publish an irrevocable pledge not to assert any patents it may have over the interoperability information against non-commercial open source software development projects.
What makes a project non-commercial? If one unemployed guy runs a project that 99 developers from Red Hat contribute to, is the project non-commercial?
"I told Microsoft that it should give legal security to programmers who help to develop open source software and confine its patent disputes to commercial software distributors and end users. Microsoft will now pledge to do so."
So, the programmers can't be sued but the users can. How much does that really help?
Almost the same story with me. The solution is to fix the bug yourself. That's great if you can pull it off, but isn't terribly practical in a lot of cases. The bug I reported probably requires tweaking a large cross section of code (every action that moves a cell, or perhaps the code that maps the data to the display). To familiarize myself with such a huge piece of source code, find the parts that need to be changed, fix them, and test, would probably take me at least 100 times longer to do than someone who wrote the code or is already familiar with it. Plus, there would be a high risk of my patch being incomplete or breaking something else due to my limited knowledge about how all of the code fits together. On top of all of that, people have already complained on a thread earlier today that the code is too messy for outsiders to work with.
While I greatly appreciate the work that is done on OOo, there does seem to be something wrong when it comes to getting bugs fixed. For example, cell notes became badly broken in oocalc 2.x because they no longer move when their cells are moved (e.g. by sorting). The bug (yes, I filed it and I am biased) has remained open for nearly two years, and the developers have classified it as an "enhancement" rather than a "defect" even though it worked fine in version 1.x and is apparently causing problems for quite a few people with no work-around. I don't mean to whine, but leaving such obvious and problematic bugs unfixed for so long isn't good for the project. I don't know if this happens because they are understaffed or if there is a problem with how things are being managed, but getting the OOo people to pay attention to bugs seems to be a problem.
You can find lists of words in various languages here: ftp://ftp.ox.ac.uk/pub/wordlists/ I don't know anything about the quality or copyright status of this.
except price is irrelevant when just doing a performance comparison. As I said, the full paper is much more informative. You may consider that extra information to be irrelevant, but that doesn't change the fact that there is a lot of info in the full paper that the submitted article doesn't even hint at. The paper, by the way, focuses on power efficiency, not performance. If people are looking at power efficiency because they want to save money on electricity (there may be other reasons to consider it, or course), then the fact that the systems themselves have very different prices seems pretty relevant to me.
OTOH, don't let me stand in the way of your fan-boyism. 75% of the computers I've bought have been Intel based. Give it a rest.
Amazingly skimpy article. No effing data whatsoever. No argument there.
I can bet a case of beer that this was run in a standard server config under Winhoze Server 2003 What kind of beer? The full paper says they were running 64-bit SUSE Linux Enterprise Server 10.
Slashdot Mirror
Yes
The "percentage of website users that are running Linux" is meaningless too, as many or maybe even most are having Firefox or Konqueror report to the server that it's IE on Windows, jusr for the websites written by idiots that redirect to a page that says "you must use IE", or are using WINE.
It's foolish to have your browser report that you're running Linux, and few GNU users are fools.
As one of the "foolish" Linux people who doesn't configure his browser to lie about running on Linux, I can say that I've encountered rather few websites worth going to that require such nonsense. If you are doing the majority of your browsing with IE on WINE, or you have your Firefox/Konqueror configured to tell every website you visit that you are running IE on Windows (rather than just the sites that really require it, which probably won't work anyway because they probably use Active-X), then you are just perpetuating the idea that it is OK to create IE/Windows-only websites.
Still meaningless. It doesn't count the half dozen or so copies I gave away.
That depends -- do the people using those half dozen copies surf the web? My point is that the 0.8% number is very similar to what I am seeing as "percentage of website users that are running Linux." If that is the measurement they are talking about, how the people got their copies has nothing to do with it. They are measuring how many people are using Linux to surf the web, not how many obtained it in any particular way.
I don't know where they get their stats from, but if I check the webserver logs for magportal.com (a fairly generic audience), I see 0.77% of unique IP addresses having "Linux" in the user-agent string over the past week.
I'm pretty sure regular contact lenses don't move when you blink, why would these?
Yes, regular contacts do move when you blink. It is most noticeable when your eyes are dry (e.g. when you are tired) because the lenses don't slide back into place as easily. Of course, if you have a weak prescription you probably won't notice the blurring when you blink because your lenses weren't doing all that much in the first place, so it doesn't matter if they are a little off. I'm not an optometrist (IANAO?), I just wear really strong contacts.
Aside from all of the other problems people have pointed out, what happens when you blink? The display moves and then settles back into position? Movement of the lens isn't a big deal when the whole thing is clear, but I would imagine it would be really annoying when there is a display on it.
Why? My sister became a freelance last year. I set her up a Debian Box with postgesql and SQL Ledger. I did the technical side. Both my dad and my sister have quite extensive accounting experience and they set the business side up in no time.
That's great if you're an accountant. If you use an external accountant, they'll want you to provide data in a format they can work with, which probably means Quickbooks for a lot of people.
curl -i 'http://discovermagazine.com/2007/dec/long-live-closed-source-software/' | head -2
I've seen claims that the ext3 filesystem updates its journal every 5 seconds. If you are running ext3 on the drives, maybe you're not losing the connection because ext3 wakes it up too often for power save to kick in.
You would think an article about ACAP would provide a link to it.
In fact, this suggests a loophole -- company X could form a separate company to create the modified version of the software, that separate company could give a copy of the modified software to company X, and company X could now use it on its intranet without offering source to its employees.
So, to get to the next page of the article I have to click the "Next Photo" link? Weird.
According to Yahoo's stock info for Google, Google has $13B in cash and it makes $4B per year in profit.
They probably incorrectly labeled all Linux Firefox users as Windows Users ... Firefox has a global 25% share (probably in excess of 90% linux browser share)
For the Linux count to be off by a factor of 100 due to not counting Firefox, Firefox would need 99% market share among Linux users. A quick and dirty analysis of a very small sample of the logs from MagPortal.com gives the breakdown (unique IP addresses with "linux" in the user-agent string [case insensitive]):
FireFox: 53%
Gecko (FireFox + Seamonkey + Minefield + etc.): 72%
Konqueror: 11%
Opera: 2%
Those numbers could easily be off by a fair amount, and I haven't excluded things like Wget or Java from the total count when computing percentages, but I would be surprised if Konqueror would come out below 1% of Linux users on a better sample.
What makes a project non-commercial? If one unemployed guy runs a project that 99 developers from Red Hat contribute to, is the project non-commercial?
So, the programmers can't be sued but the users can. How much does that really help?
While I greatly appreciate the work that is done on OOo, there does seem to be something wrong when it comes to getting bugs fixed. For example, cell notes became badly broken in oocalc 2.x because they no longer move when their cells are moved (e.g. by sorting). The bug (yes, I filed it and I am biased) has remained open for nearly two years, and the developers have classified it as an "enhancement" rather than a "defect" even though it worked fine in version 1.x and is apparently causing problems for quite a few people with no work-around. I don't mean to whine, but leaving such obvious and problematic bugs unfixed for so long isn't good for the project. I don't know if this happens because they are understaffed or if there is a problem with how things are being managed, but getting the OOo people to pay attention to bugs seems to be a problem.
Just two weeks ago:
Sberbank (Savings Bank Of The Russian Federation) Chooses SCO's OpenServer 6 to Upgrade Thousands of Servers.
They must be feeling really good about that decision now.
You can find lists of words in various languages here:
ftp://ftp.ox.ac.uk/pub/wordlists/
I don't know anything about the quality or copyright status of this.
except price is irrelevant when just doing a performance comparison.
As I said, the full paper is much more informative. You may consider that extra information to be irrelevant, but that doesn't change the fact that there is a lot of info in the full paper that the submitted article doesn't even hint at. The paper, by the way, focuses on power efficiency, not performance. If people are looking at power efficiency because they want to save money on electricity (there may be other reasons to consider it, or course), then the fact that the systems themselves have very different prices seems pretty relevant to me.
OTOH, don't let me stand in the way of your fan-boyism.
75% of the computers I've bought have been Intel based. Give it a rest.
Amazingly skimpy article. No effing data whatsoever.
No argument there.
I can bet a case of beer that this was run in a standard server config under Winhoze Server 2003
What kind of beer? The full paper says they were running 64-bit SUSE Linux Enterprise Server 10.