It displays documents in a device independent and resolution independent form. It is also an open standard. If you need a print-ready electronic document to look the same in most places, PDF is a reasonable option.
XML alone doesn't solve this need (nor are there really compelling schemas which can replace PDF), which is one reason why there are XML->PDF tools such as FOP.
FWIW: I think Adobe does use XML for their PDF forms these days.
xpdf is a collection of some of my favorite PDF utilities & is F/OSS. There are de-drm patches in the wild & some distributions (such as gentoo) allow you to apply them fairly painlessly. For people with X, you can use a viewer. Any platform should be able to use programs which convert the PDF to PostScript (which may then be converted back to an un-DRMed PDF with other software).
But he is right: Mozilla/Firefox probably didn't benefit Netscape. The company.
They open sourced it in 1998. Months later, they were bought by AOL & left to stagnate as a company. Their self-named browser is still unpopular. Fortunantely the open source derivatives aren't.
Netscape-derived browsers would probably be dead were it not for open source. They might have continued to cling at the niche of a few corporate deployments (as it was free, fairly easy to remotely administer, and was not IE), but that's it. There would be fewere fans, fewer promotion, and fewer users. They would not have put out anything like Firefox, which required some minimal level of community.
So open source proved good for users. Possibly good for the internet. It wasn't necessarily good for AOL-Netscape.
However, nothing is stopping me from writing my own client libraries under whatever-the-fuck-i-want license and using those instead of MySQL's license.
Right...But the cost of doing so would probably be greater than the cost of a license. It would take a lot of work to make them as easy to use as those build by MySQL AB & presumably MySQL AB is in the best position to make them stable & efficient.
Most who really care about distributing proprietary software charge money to their customers & can pass the license cost of a database off onto the customer. The lowest risk option would be to NOT develop your own libraies.
Of course, if they can't justify the cost, they should just use postgres or some other database whose license conforms to their needs.
For those committed to using MySQL clients for free in a proprietary app, the best bet is probably the legacy "public domain" version (I suspect these are the old LGPL libraries, but haven't looked more closely).
This means I can write a completely commercial MySQL app, and not GPL any of it. (If you disagree, remember that connecting to a GPL'd webserver from IE doesn't require MS to open IE's source code.)
That is how I read the GPL, but IANAL & that isn't how MySQL AB (who has lawyers) reads the GPL. They claim
If you include the MySQL server with an application that is not licensed under the GPL or GPL-compatible license, you need a commercial license for the MySQL server.
If you develop and distribute a commercial application and as part of utilizing your application, the end-user must download a copy of MySQL; for each derivative work, you (or, in some cases, your end-user) need a commercial license for the MySQL server and/or MySQL client libraries.
I personally don't think it would be worth the risk to test this & would be conservative. If I was making proprietary software, I'd either pony up for the license or choose another database.
If you include the GPL version for free, you must comply with the GPL same as any other piece of GPL code, that is give away the source code too..
To be more precise, it is more lenient than the GPL alone: if you distribute your software with the MySQL client libraries, you can use any OSI-approved license, even if it is not GPL-compatible.
So if your using MySQL unmodified, you just give a copy of the source tarball, easy.
You must include your source code. If MySQL was released under the LGPL (which it isn't), then you'd be right--you'd only care if you modified MySQL. But is is released under the fully copy-left GPL, which forces anything that liks to the client libraries under their license. That is: If you use ODBC, JDBC, or the C MySQL libraries in your application, you must release it under an OSI-approved license or buy a fully copy of MySQL. There is no debate about this. You are using their libraries and must follow the GPL.
The free commercial databases may let you distribute the binaries for free (or they may not, not sure of the licensing terms) but you certainly can't modify them, so you don't gain anything you wouldn't already be able to do with MySQL.
As above, if they allow distribution of your application under any license of your choosing, they do allow something the GPLed MySQL doesn't.
After all, if your distributing an unmodified copy of MySQL why would distributing the sourcecode hurt you? people can just download it from mysql.org anyway.
Because you must distribute the full source to your entire application, which people couldn't download! I called a few people on this. They were using the GPLed-version of MySQL for their proprietary apps. I told them we wanted them to follow the license--open source their app, use a different database, or buy the commercial license. They invariably bought a license.
The gratis version of MySQL is released under the GPL with the exception that it can be included in ANY F/OSS software. If you are creating an open source product or won't distribute your software, this is great. If you are producing commercial proprietary software, you must purchase a license to MySQL. I believe that many of the gratis, proprietary programs don't carry this restriction.
OO.org is dog slow Linux, faster on windows. but still slower then Excel.
note 90% of the time I need a Spread sheet I'm in Linux and use OO.org any way.
OO.o on my Linux is faster than MS Office on my Windows on the same machine.
Gnumeric is even better (more featureful & those features WORK) than OO.o Calc & is faster still, so that is what I use. The win32 port has come a very long way. It isn't as good as the Linux version, but I find I use it at least as often as MS Excel. You might give Gnumeric a try on Linux!
I elsewhere commented on how poor the PowerPoint Viewer is (OO.o is FAR better).
There is less demand for a simple player for OO.o impress, as you can get the full-blown Impress for the same price (and both diskspace and bandwidth are relatively cheap). Being able to export to PowerPoint format also means you can use the PowerPoint Viewer (though, as I said, you wouldn't really want to).
There used to be a StarOffice player, but it was discontinued. I suspect that clever people could just write XSLT transforms to dump your presentation into whatever stand-alone format you want.
The "Export to PDF" feature is quite useful to make read-only, portable copies of any OO.o-generated content (though you may lose embedded animations).
Office applications aren't an essential component on our analysis machines at work & we choose to run OO.o on all of them for the rare time someone needs to use them. We recently had a meeting with the company that writes our analysis software & wanted to show both live demos and presentations on one box. Because all of the people from that company & many of our own people were running MS Office on their own machines, I thought temporarily installing the gratis MS PowerPoint Viewer would be a wise move--I had heard of compatibility problems between MS Office & OO.o and these would be intricate presentations with embedded graphics and equations.....
All of which the PowerPoint Viewer completely choked on, leaving boxes with red Xs through them. Since OO.o was already on the machine, I opened the presentations in it & they all looked great!
There's a difference in me implying something and you assuming it. I didn't imply anything you accused me of implying. That you resort to accusing me of making implications indicates, at least, I made no direct statement about the lack of details in the article which you can prove "wrong."
I pointed out that these had been addressed in the article
You said that the presence of what you thought were answers was evidence that I hadn't read the article.
If you had posted based on what you have subsequently tried to imply you meant...
I'm sorry for trying to clarify my original post!
then I'd simply point out that trying to misapply standards meant for scientific results on material that is not presented as a scholarly article.
I'd say "fine" -- but I was still curious as to the answers! God forbid we learn something beyond what was in the article!
This is intellectual arrogance, simply designed to karma whore.
First you accuse me of "implying" things & now of whoring for meaningless points?
Assume that the editors of Nature were prepared to open themselves up to complaints of professional misconduct which may well threaten their standings on the editorial board on the basis of a trivial comparison between Wikipedia and the Britannica.
Misconduct?! There's a difference between "sloppy" science & actual misconduct.
Then what motivation do you ascribe for them doing this?
Results can be biased without a conscious motive. Why do you think they need to be malicious to run a sub-par "experiment?"
I'm also amused that you assume that the editors of an extremely well-respected science journal wouldn't follow standard experimental design and methodological procedures reinforced ad nauseum from the first days of university training.
Why? The sample size is small; the article isn't peer-reviewed or intended to be "serious;" it was an editorial. It is a preliminary study at best & I have seen first-hand the short-cuts that are made.
But then again, I don't have a superiority complex.
Your bickering and accusations of "whoring" have really shown that!
then I'm sure you emailed Nature asking for details of their methodology.
No. I asked on here. I then waited until someone posted a link to a blog article with links to more details.
If you had and then posted their reponse, then that would have deserved an Insightful rating.....you didn't deserve an Insightful.
You certainly seem obsessed with karma! I could care less.
Leave physics and try law
To be quite arrogant, I'd rather scientists have curiosity & you don't seem to share mine. Perhaps you can become a bureaucrat? They get to call each other worse things than "karma whore."
Am I dreaming or are you now saying that the article answered the questions, just that these answers were not as informative as you'd prefer?
You're not dreaming. I conceded that you might be happy with the puff piece, but that (despite your claims), I did read the article & that I wasn't completely satisfied.
Which is not what you originally imply in your post, which was that they weren't addressed at all.
My exact words were:"The article lacks important details." I stand by that.
I think it matters little how they were chosen: whether they were pet topics of one of the editors at Nature or by some other method is irrelevant to the comparison.
It does matter. People are biased. If articles were selected by people who read both articles & knew the source, the samplings could be tainted to favor one of the sources.
As to the third question: the article states that the reviewers were unaware of the source of the articles.
Again, it stated they "were not told which encyclopedia the stories were from" (emphasis mine). If they were emailed or faxed or mailed a copy of the two articles & were told one came from Wikipedia & one came from Britannica (but not which came from which), it wouldn't take a rocket scientist to find which article came from which source. Even if they weren't informed of the original sources, I think it should be simple for professionals who routinely perform literature searches to figure it out.
One would assume that the editors from Nature are aware of the value of ensuring their data are valid.
The blind could be established if the person who copied the articles was different than the one who described the task to the scientists & who analyzed the data. The sources could be revealed after the results were in.
A news-bite is hardly worthy of publishing rigourous data about the methodology
Ah...we agree on something. The article didn't detail methodology well & I had just wanted more details. (Nowhere, by-the-way, did I say the article shouldn't have been published or that it should have answered all of my quesions in more depth. I only said I was left with those questions.)
...you are required to meet deadlines imposed by customers and to produce end products specified by customers, otherwise your business fails...you'll naturally assume that... software engineers working on customer-driven projects are the minority. Sad to say, that ain't the case.
Stallman would say that you shouldn't take that kind of software engineering job. Ideally, I suppose, you'd find a job where you can use or develop only Free software. But Stallman has said before that he realizes that this can't always be done, and pointed out that there are many jobs out there that aren't centered around software development that you could make a living without relying on proprietary software. And he is right about that--many people do get by with minimal use of computers today.
It is unlikely that someone who is most passionate about and most talented at software engineering would choose a career in another field. But if you shared RMS's sense of ethics (which, admittedly, most people don't), you probably should choose your conscience first. A trite example: if you were morally opposed to weapons & were an engineer who was offered a job making weapons, would you?
I have my own list of F/OSS projects
on
Season's Givings?
·
· Score: 1
All your questions are answered in the text of the article.
Are we reading the same article? They weren't answered very well at all.
How were these chosen?
The article said they were science-related topics selected by Nature that covered things like "agent orange, quarks, and synchrotrons." This doesn't speak to how these were selected. Favorite topics by a person or people at Nature? Randomly drawn from a scientific glossary of some sort? Somehow automatically selected based on most significant topics in wide-reaching fields?
Were they on the same topics for the two encyclopedias?
I'll agree that the article implies they were.
Was this done double-blind?
The article does not directly answer this. It claims they weren't told which encyclopedia they came from. Were they told that one came from Wikipedia & one came from EB? If so, was the reviewers sufficiently isolated that they couldn't go back to the sources to see which the article came from (and thus bias results)?
They looked at 42 articles. How were these chosen? Were they on the same topics for the two encyclopedias? Was this done double-blind?
the difference in accuracy was not great: the average science entry in Wikipedia contained around four inaccuracies; Britannica, around three.
About a 30% difference (running their numbers of 162 and 123), which is big. More importantly (from the article):
Only eight serious errors, such as misinterpretations of important concepts, were detected in the pairs of articles reviewed, four from each encyclopedia.
The article also doesn't comment if corrections were made to the Wiki and submitted to the editors of Brittanica....
I do the same thing on *nix boxes, including OS X. These platforms aren't as exploited. Period. I do still run ClamAV on public mail servers or Samba boxes which allow upload. It won't hurt my computer, but I don't want to have other computers hurt by my box either.
Um, what about with a manual scan? I do much the same as the other guy: I run Win2k, and I don't keep any AV, anti-spyware, or firewall software running. Every once in a while, I download the latest version of Spybot or AVG and let it have a look.
By "automated scan." I do not mean real-time protection. I mean that you are using a piece of software to look for you. And you are. I would say you are using antiviral and anti-spyware software.
Why waste processor cycles on buggy and unstable "protection" software when safe practices are enough, and their success can be confirmed with occasional checkups?
Checkups are fine. And if you cron them to run once a week or once a month, so much the better. (I, by the way, think it is better to waste an insignificant amount of harddrive space keeping the non-realtime scanners installed.)
You make a good point re. balance. None is too little. That was my only point.
I run no antivirus or anti-spyware software, mainly for performance reasons.
Real-time scanners can, indeet, hurt performance. Most people wouldn't notice a performance hit in most applications & the scanners can be temporarily suspended for applications which are exceptions. A non-realtime scan can be done at any time when performance doesn't matter & would provide you with some measure of safety.
I haven't gotten any spyware in a long time
I would be surprised if the aggressive adware/spyware programs agreed with your self-assessment. But, then, how would you know you had spyware without an automated scan? Most successful spyware is covert.
and (to my knowledge) have never been infected by a virus/trojan/worm...
Again, "ignorance is bliss" is not the same as "I know I have no problems."
I don't open attachments
In what job can you afford to ignore everything sent to you?
If you ever have to download unsigned files (and most people do), I'd AT LEAST have a proxy with ClamAV installed.
Even for the knowledgable, it is wreckless not to use some measure of protection on a habitually-exploited system.
One solution (somewhat) has been http://wikipedia.com/. I know this is not a dictionary but works if you need to double check a spelling, but mainly I have found it useful while writing scientific pieces to double check a few pathways or cell types.
AbiWord is a capabale F/OSS word processor which is available for most platrforms. It is lighter than OO.o Writer (though that also means it lacks SOME of OO.o's features). One of the really nice features is that it supports a number of plugins. There are plugins which allow you to search a selected word on Wikipedia, google, and dict.org. Also, on *nix, it can use GDict.
just around the time the first year of MAKE is almost over -- translate to: we have to all renew
Ah--someone didn't use the get 1 free issue coupon;-)
there is a shashdot post from a MAKE contributor.
More insiduously, Phillip Torrone is associate editor of the magazine.
At least he picked a good title for the slashdot post.
And, in addition to being a not-so-subtle marketing ploy, the list does have some good gadgets. And it does seem interesting to geeks: highly ranked on digg & del.icio.us.
I used to run my own consultancy and I used OpenOffice (well StarOffice back then) as my office suite. I found it more useful to send documents around in PDF format instead of sending word documents to my clients. Most of my clients could read PDF back then (this was '99/2000) - even more should be able to now.
You should be happy to know that OO.o compatibility with MS Office format has improved in the last 5 years. I have a number of difficult testcases--100 MB theses & the like. OO.o will sometimes take longer to open them. Once open, some odd macro-dependent additions don't work. But nearly everything else does. Most importantly, it can write DOCs which seem to be more readable by both MS Office and OO.o.
I thought the same thing. I needed to write an important document, which would be passed around with MS Word's change tracking, comments, etc. I reluctantly used MS Office 2003 to generate the monstrosity & it was a nightmare getting the document to work with a copy of Office 97 that a colleague was using. I tried exporting it in all of the previous versions of the DOC format that MS Office 2003 let me choose from. No luck. I ended up opening it in OO.o Writer. It was slow. There was some stuff that wasn't ideal. But it opened. And, better, when I saved it, it opened & looked good in Office 2003 AND Office 97.
I primarily use abiword and OO.o writer on linux, so this might be a case of PEBKAC.
However, I've seen this sort of thing before--OO.o will often struggle opening a file, but typically once I write a file with it, it CAN be read by both OO.o and MS Word.
Slashdot Mirror
It displays documents in a device independent and resolution independent form. It is also an open standard. If you need a print-ready electronic document to look the same in most places, PDF is a reasonable option.
XML alone doesn't solve this need (nor are there really compelling schemas which can replace PDF), which is one reason why there are XML->PDF tools such as FOP.
FWIW: I think Adobe does use XML for their PDF forms these days.
xpdf is a collection of some of my favorite PDF utilities & is F/OSS. There are de-drm patches in the wild & some distributions (such as gentoo) allow you to apply them fairly painlessly. For people with X, you can use a viewer. Any platform should be able to use programs which convert the PDF to PostScript (which may then be converted back to an un-DRMed PDF with other software).
So obviously He must e a *BSD user.
But he is right: Mozilla/Firefox probably didn't benefit Netscape. The company.
They open sourced it in 1998. Months later, they were bought by AOL & left to stagnate as a company. Their self-named browser is still unpopular. Fortunantely the open source derivatives aren't.
Netscape-derived browsers would probably be dead were it not for open source. They might have continued to cling at the niche of a few corporate deployments (as it was free, fairly easy to remotely administer, and was not IE), but that's it. There would be fewere fans, fewer promotion, and fewer users. They would not have put out anything like Firefox, which required some minimal level of community.
So open source proved good for users. Possibly good for the internet. It wasn't necessarily good for AOL-Netscape.
Most who really care about distributing proprietary software charge money to their customers & can pass the license cost of a database off onto the customer. The lowest risk option would be to NOT develop your own libraies.
Of course, if they can't justify the cost, they should just use postgres or some other database whose license conforms to their needs.
For those committed to using MySQL clients for free in a proprietary app, the best bet is probably the legacy "public domain" version (I suspect these are the old LGPL libraries, but haven't looked more closely).That is how I read the GPL, but IANAL & that isn't how MySQL AB (who has lawyers) reads the GPL. They claimI personally don't think it would be worth the risk to test this & would be conservative. If I was making proprietary software, I'd either pony up for the license or choose another database.
Read more about MySQL licensing.
The gratis version of MySQL is released under the GPL with the exception that it can be included in ANY F/OSS software. If you are creating an open source product or won't distribute your software, this is great. If you are producing commercial proprietary software, you must purchase a license to MySQL. I believe that many of the gratis, proprietary programs don't carry this restriction.
Gnumeric is even better (more featureful & those features WORK) than OO.o Calc & is faster still, so that is what I use. The win32 port has come a very long way. It isn't as good as the Linux version, but I find I use it at least as often as MS Excel. You might give Gnumeric a try on Linux!
I elsewhere commented on how poor the PowerPoint Viewer is (OO.o is FAR better).
There is less demand for a simple player for OO.o impress, as you can get the full-blown Impress for the same price (and both diskspace and bandwidth are relatively cheap). Being able to export to PowerPoint format also means you can use the PowerPoint Viewer (though, as I said, you wouldn't really want to).
There used to be a StarOffice player, but it was discontinued. I suspect that clever people could just write XSLT transforms to dump your presentation into whatever stand-alone format you want.
The "Export to PDF" feature is quite useful to make read-only, portable copies of any OO.o-generated content (though you may lose embedded animations).
Office applications aren't an essential component on our analysis machines at work & we choose to run OO.o on all of them for the rare time someone needs to use them. We recently had a meeting with the company that writes our analysis software & wanted to show both live demos and presentations on one box. Because all of the people from that company & many of our own people were running MS Office on their own machines, I thought temporarily installing the gratis MS PowerPoint Viewer would be a wise move--I had heard of compatibility problems between MS Office & OO.o and these would be intricate presentations with embedded graphics and equations.....
All of which the PowerPoint Viewer completely choked on, leaving boxes with red Xs through them. Since OO.o was already on the machine, I opened the presentations in it & they all looked great!
It is unlikely that someone who is most passionate about and most talented at software engineering would choose a career in another field. But if you shared RMS's sense of ethics (which, admittedly, most people don't), you probably should choose your conscience first. A trite example: if you were morally opposed to weapons & were an engineer who was offered a job making weapons, would you?
These are the projects that have been worthy of my money. There are donation links and methods of payment listed:
http://arc.nucapt.northwestern.edu/F/OSS
They looked at 42 articles. How were these chosen? Were they on the same topics for the two encyclopedias? Was this done double-blind?
I do the same thing on *nix boxes, including OS X. These platforms aren't as exploited. Period. I do still run ClamAV on public mail servers or Samba boxes which allow upload. It won't hurt my computer, but I don't want to have other computers hurt by my box either.
You make a good point re. balance. None is too little. That was my only point.
If you ever have to download unsigned files (and most people do), I'd AT LEAST have a proxy with ClamAV installed.
Even for the knowledgable, it is wreckless not to use some measure of protection on a habitually-exploited system.
At least he picked a good title for the slashdot post.
And, in addition to being a not-so-subtle marketing ploy, the list does have some good gadgets. And it does seem interesting to geeks: highly ranked on digg & del.icio.us.
I thought the same thing. I needed to write an important document, which would be passed around with MS Word's change tracking, comments, etc. I reluctantly used MS Office 2003 to generate the monstrosity & it was a nightmare getting the document to work with a copy of Office 97 that a colleague was using. I tried exporting it in all of the previous versions of the DOC format that MS Office 2003 let me choose from. No luck. I ended up opening it in OO.o Writer. It was slow. There was some stuff that wasn't ideal. But it opened. And, better, when I saved it, it opened & looked good in Office 2003 AND Office 97.
I primarily use abiword and OO.o writer on linux, so this might be a case of PEBKAC.
However, I've seen this sort of thing before--OO.o will often struggle opening a file, but typically once I write a file with it, it CAN be read by both OO.o and MS Word.