While this is probably a dig at the cost of the project, many now say the 'i' stands for 'independent' rather than the original 'inexpensive.' There are good reasons to buy very large, high quality SCSI disks for a project like this.
"If you want to help, paypal some money to slash at peereboom dot us, or you can use the OpenBSD ordering system. Be sure to mention its for the cvs machine."
The price of a fully populated PowerVault 220S with 4 hour on-site warranty is
about $12500 USD including tax and shipping. Let me kick this off by donating $250 to this cause. A person that shall not be named donated a PERC4/DC RAID controller.
Stacker was actually illegally included in DOS 6.0 as DoubleSpace. See here or here for an english translation. They removed it for 6.2 & created DriveSpace for 6.22. DriveSpace was used in almost an identical manner, but was incompatible wiht DoubleSpace, causing upgraders who were using drive compression no end of grief.
Open Tax Solver is the only F/OSS tax program worth mentioning. It is better than doing it by hand but (if you are used to handholding from TaxCut, TurboTax, and similar products) you will need to be ready for a shock. It is under active development & started out as merely a simple calculator. You would feed in a text file of what numbers you would put on which lines & it would spit out what to put on all of the other lines. So you still need to be familiar with how to do your taxes by hand--you just don't need to have a calculator when you do this. The advantage of this is that it is very flexible--the same program can and is being used for state and other taxes than the US Federal 1040. The disadvantage, of course, is that you need to know a little something & be able to edit that text file.
Someone has since developed a GUI for it, but it is still quite new & somewhat untested. I haven't a clue if the GUI is as flexible as the CLI program.
The output is a textfile. They suggest you sit down with the text file open & fill out a fillable PDF form by hand. Acroread 7 supposedly supports filling in form data from a text file, so that will be the next big improvement to OTS. The catch is you still have to print out the form & mail it in. I don't know how likely efile will be--just as with the open source personal finance programs downloading bank statements, there is generally a lack of information sharing unless you are Intuit or H&R Block.
Don't like this? Then use a free (as in beer) web service through freefile. They list dozens of sites where you can complete and efile federal and some state taxes. Most allow you to keep a PDF of the filed forms for your own records or for a snailmail submission.
It bears mentioning that RAR was designed from the get-go to support the features mentioned, and has for what, seven years now?
RAR was not designed from the get-go to support Reed-Solomon repair. This was a feature added for 2.0 (released 1996). So, it has supported it for 8 years, but it was also back-ported.
You also don't need to use RAR to use the same file recovery mechanisms. Use PAR on ANY type of file to benefit from it!I'm not trying to trash 7-zip, but I'm also not going trying 7-zip when RAR has the track record it does for doing what it was designed to do.
7-zip does what it is designed to do as well: use an open source program to both create and extract archives with very high compression ratios.
We would like to thank the OpenSSH community for their continued support to the project, especially those who contributed source and bought T-shirts or posters.
You can also do what I plan to do: donate surplus hardware to OpenBSD, which runs the project. OpenBSD accepts other donations too:checks, credit cards, paypal.
7zip? Unmature? It is perfectly usable. I finally deleted my licensed copy of Winzip because 7zip does its work better. 'nuff said.
7-zip is also on the win32 boxes I administer & winzip isn't. But this kind of use is not enough to prove the kind of maturity that I'm talking about. One piece of evidence is that most users aren't using it for 7zip archives. Another is that the *nix version of 7-zip is about 8 months old and listed as beta. Only in October did KDE and GNOME add support IN THEIR CVS. So, no, I wouldn't release any archives to the world in the format unless I had a significantly better technological reason to use it over bzip2.
the RAR compression algorithm, which is proprietary
And there lies the problem that I mentioned. The compression algorithm isn't Free.
So the guy is basically paranoid about keeping his "trade secret" a secret, which makes perfect sense from a business perspective.
But why bother catering to his business when you could use something like 7-zip?
As far as FOSS is concerned, he even presents decompression code free for all to use.
It also makes sense from a business perspective: formats which are distributed gain you more customers when they can be read by everyone. MS makes a free (as in beer) viewer for their office formats too.
How much of a problem is a proprietary compressor if the end-user never has to deal with it?
When there is a completely Free and Open format that works as well or better, it is a significant problem. I know many developers who don't have a RAR program on their system. They never write and never need to read the format. I, myself, use an LGPLed decompressor. But I don't have it on all of my boxes & even uninstall it on some of them after I use it.
Perhaps you are right & only a zealot would take the position I take. But test data benefits developers & you should be using WHATEVER archival format they use for the other software that they use and make. I don't think this will be RAR or another proprietary format.
Of course I'm sure there are plenty of unpopular Open formats that would also be unsuitable too. Your primary concern should be what everyone else is using. The license of your archiver has already impacted this, so perhaps we don't need to argue about how (un)Free RAR is.
I see it like gzip: gzip is defined by a couple of RFCs, and it's just a file format and a specification for decompressing the data. As long as you end up with something in.gz format
Well, where is the RFC or open documentation for the RAR format? Where are the free compressors? And where is the software distributed using this format? RAR is much less Free than gzip, bzip2, or 7-zip.
But RAR/PAR encoding use a Reed-Solomon error correction scheme, so you can send a few additional files and the data blocks within them can be used to replace *any* lost data blocks from the original set. It's really crafty.
You're right. It is clever & is better than md5 for repair purposes. Which is why is aaid md5 was probably enough for test data (I think bz2 archives are the way to go, so I'm a bit biased).
But, once again, this is not a unique feature of rar. I know I've seen proprietary ZIP programs that offer it. If 7-zip doesn't currently offer it, it will in the future (I know I saw it on the roadmap & think I saw an option the last time I used 7-zip on win32).
I acknowledged this in another post. I also use an LGPLed unrarer. But the RAR compression algorithm is, as the license makes very clear, "proprietary." I've seen very little F/OSS distributed as RAR archives, and I don't think it is coincidence.
While there are Free rar unpackers, the primary packer/unpacker has a proprietary license. He is catering to open source developers. It is a poor choice.
md5 probably provides enough integrity checking for test data & split/cat make splitting/reassembling ANYTHING easy.
RAR doesn't have a monopoly on integrating these, though. 7zip certainly has many of the service features available in rar.
I'm starting up a free data site to provide test data for anything you can imagine: images for compression and format interpretation, text and audio for language processing, programming language examples to test parsing, and more.
Your audience will be developers. Hopefully F/OSS developers. So distribute in a developer-friendly format. BZ2 files can be decompressed with Free software & most of the proprietary applications out there that would be decompressing alternative formats.
While 7-zip is a nice open format & archivers come with open licenses, it isn't mature enough yet. The *nix ports are still fairly young. And few third-party apps would unzip them. Finally, RAR does not use an open license. I would discourage you from using it for this project.
A big problem is that a majority of the journals are run by for profit companies like Elsevier. They get free content (from tax-funded scientists), cheap editing (through free peer review), and sell subscriptions for huge sums (to tax-funded libraries). They have a monopoly and, worse, they are taking money from a notoriously unfrugal consumer: government entities that don't bid.
Publishing, like research, is a public good. It can and should be run by the state. Universities and national labs already have small-run publications which they give away for free. I would bet that, even for these government run institutions, it would be more cost-effective to self-publish. They could put those parts of academic publishing that cost the most (typesetting, printing, mailing) up for bid.
The only tricky point is getting scientists to publish to the new journals--prestige of the journal is quite important.
I'm sure Sun would be willing to address each of these points if the government had bought into Star Office (which uses much of the OO.o codebase). There are also independent support providers which would write out a contract & provide training for OO.o.
Reveal codes is an absolutely wonderful feature for fixing broken documents. Not everyone uses Word styles (I'm tempted to say a minority do) & you WILL get broken, kludgy documents. If for no other reason than this, it would be nice to see where codes start/stop.
It is nice to see exactly where an image is anchored or when a hyphen/spacing is breaking/nonreaking and when these or line/page breaks are optional or forced.
It is also extremely useful to see when a STYLE starts/stops! Third-parties sell an atrocious hack to put a reveal codes feature into Word. The real thing is better.
It is the next best feature to using transparent plaintext formats like docbook/LaTeX, where you can get the same info.
While acroread 5 is horrible (and the parent therefore deserves the Funny mod), the beta version of acroread 7 is a nice enough GTK app. I still have complaints & it isn't enough to switch me off of xpdf, I no longer cringe when I need some peculiar features from acroread.
ck is a nice patchset. For something with slightly more, CKO offers everything ck does with Reiser4, Supermount, Alan Cox's -ac patchset, software suspend, updates to libata/ALSA/Bttv, and more.
If anyone knows how to donate small amounts of money to the developers, please let me know: both ck and cko are on my list of projects to eventually donate to (linked to from my URL).
arXiv http://www.arxiv.org/ usually contains the.tex source for any preprint published there.
Great point!
The original author of LaTeX has a very great introduction book (it has a blue and yellow cartoon cover...).
Lamport's LaTeX: A Document Preparation System. It is a bit dated now (it is over 10 years old!) & I would suggest the more recent (and, IMHO, more polised) LaTeX Companions instead. But it is still a good text and, with Knuth's works, provide interesting gems that only people who have thought so much about the low level stuff can provide. I agree it makes a better intro than a reference.
If you want to benefit from it without learning it, you can use a number of GUIs. Scientific Workplace on win32 (commercial, but good to push on those using Word) or LyX (F/OSS) for nearly any platform or many others. Even abiword can write LaTeX!
It isn't difficult to learn & becomes much more powerful when you eventually ditch the GUI & either use a quality TeX-focused editor like KILE (KDE), TeXnicCenter (win32), TeXShop (OS X) (all F/OSS) or your favorite multi-purpose editor. I prefer vim with LaTeX-Suite.
The best way to learn is to look at other code. Either get some from peers, from the net, or make some in either the GUIs or the friendlier editors. Then just write.
If you need a reference, you can usually learn to google for how to do something (or post to comp.text.tex). I maintain a list of www links. You might find something useful, but I can't suggest the best starting point from that list. The best introductory book I've used is Guide to LaTeX. The other books in LaTeX Companions are also excellent for reference, particularly The LaTeX Companion.
LaTeX is on ALL popular platforms, including win32, OS X, Linux, and *BSD
and Excel.
Works on win32 or OS X or throgh wine. Or you can use gnumeric/OO.o (which I find better) or a REAL plotting software. One of the OTHER advantages of using a real plotting packages such as those I listed is being able to export into a LaTeX-friendly graphics format.
The only thing that Windows has going for it is MS Office but since you can get it for OS X that argument is mute. If you want KDE or Gnome you can run it inside X11 on OSX. Many Linux apps will compile natively and you may not need X11 at all.
And you can usually use VirtualPC. Yes--OS X provides passable access to most programs. But passable!=superior. I would rather be able to use all the software I run on an OS that ran them as cleanly as possible, rather than know that my OS could also run a bunch of stuff I don't actually use.
Nearly all the programs I use are F/OSS, so I use F/OSS operating systems. They just seem to work better. Even if I had a Mac, I'd probably need to use a PowerPC version of Linux.
Personally it is not so much the operating system as the window manager.
This can go a long way, but switching between applications less can actually make a lot of people more productive, so you may overstate the importance
I use fluxbox becase I like being able to scoll between virtual desktops with my mouse scroll wheel.
Fantastic feature, but this is hardly unique to Fluxbox.The advantage of Linux is that you have tons of window managers to choose from, as opposed to Windoze of OSX where you are limited to the one provided.Simply not true. Not only can OS X run X & windowmanagers on top of it, but even on win32 you have alternatives like bb4win (blackbox for windows).
As a scientist, where I do most of my work in MS Office...I basically have to use MS Office because I need to interoperate with my peers and coworkers.
This is sad, but true. If I am primary author, I do it in LaTeX & get it done in a tenth of the time. But people are locked into Word & Powerpoint and my life is occasionally made a little more painful because of that. OO.o and abiword go a long way, as does latex2rtf. Depending on how much content I am creating, it is often faster to use my preferred tools: LaTeX and vim.
Furthermore, Excel (every scientists best friend), is still far and away the best spreadsheet application and to me is Window's so called "killer app".
While Excel is a fine enough spreadsheet (I can't think of anything I like from it that Gnumeric and OO.o don't do), most scientists need much more than a spreadsheet. They need an industrial strength plotting program, a'la Microcal Origin, Kaleidagraph, grace, gnuplot, Matlab, Igor, hippodraw, etc. It isn't my best friend & even the people who are stuck on Word that I collaborate with discourage anyone from using Excel for anything other than quick & dirty.
Slashdot Mirror
While this is probably a dig at the cost of the project, many now say the 'i' stands for 'independent' rather than the original 'inexpensive.' There are good reasons to buy very large, high quality SCSI disks for a project like this.
"If you want to help, paypal some money to slash at peereboom dot us, or you can use the OpenBSD ordering system. Be sure to mention its for the cvs machine."
Stacker was actually illegally included in DOS 6.0 as DoubleSpace. See here or here for an english translation. They removed it for 6.2 & created DriveSpace for 6.22. DriveSpace was used in almost an identical manner, but was incompatible wiht DoubleSpace, causing upgraders who were using drive compression no end of grief.
Open Tax Solver is the only F/OSS tax program worth mentioning. It is better than doing it by hand but (if you are used to handholding from TaxCut, TurboTax, and similar products) you will need to be ready for a shock. It is under active development & started out as merely a simple calculator. You would feed in a text file of what numbers you would put on which lines & it would spit out what to put on all of the other lines. So you still need to be familiar with how to do your taxes by hand--you just don't need to have a calculator when you do this. The advantage of this is that it is very flexible--the same program can and is being used for state and other taxes than the US Federal 1040. The disadvantage, of course, is that you need to know a little something & be able to edit that text file.
Someone has since developed a GUI for it, but it is still quite new & somewhat untested. I haven't a clue if the GUI is as flexible as the CLI program.
The output is a textfile. They suggest you sit down with the text file open & fill out a fillable PDF form by hand. Acroread 7 supposedly supports filling in form data from a text file, so that will be the next big improvement to OTS. The catch is you still have to print out the form & mail it in. I don't know how likely efile will be--just as with the open source personal finance programs downloading bank statements, there is generally a lack of information sharing unless you are Intuit or H&R Block.
Don't like this? Then use a free (as in beer) web service through freefile. They list dozens of sites where you can complete and efile federal and some state taxes. Most allow you to keep a PDF of the filed forms for your own records or for a snailmail submission.
Perhaps you are right & only a zealot would take the position I take. But test data benefits developers & you should be using WHATEVER archival format they use for the other software that they use and make. I don't think this will be RAR or another proprietary format.
Of course I'm sure there are plenty of unpopular Open formats that would also be unsuitable too. Your primary concern should be what everyone else is using. The license of your archiver has already impacted this, so perhaps we don't need to argue about how (un)Free RAR is.Well, where is the RFC or open documentation for the RAR format? Where are the free compressors? And where is the software distributed using this format? RAR is much less Free than gzip, bzip2, or 7-zip.
But, once again, this is not a unique feature of rar. I know I've seen proprietary ZIP programs that offer it. If 7-zip doesn't currently offer it, it will in the future (I know I saw it on the roadmap & think I saw an option the last time I used 7-zip on win32).
I acknowledged this in another post. I also use an LGPLed unrarer. But the RAR compression algorithm is, as the license makes very clear, "proprietary." I've seen very little F/OSS distributed as RAR archives, and I don't think it is coincidence.
While there are Free rar unpackers, the primary packer/unpacker has a proprietary license. He is catering to open source developers. It is a poor choice.
md5 probably provides enough integrity checking for test data & split/cat make splitting/reassembling ANYTHING easy.
RAR doesn't have a monopoly on integrating these, though. 7zip certainly has many of the service features available in rar.
While 7-zip is a nice open format & archivers come with open licenses, it isn't mature enough yet. The *nix ports are still fairly young. And few third-party apps would unzip them. Finally, RAR does not use an open license. I would discourage you from using it for this project.
A big problem is that a majority of the journals are run by for profit companies like Elsevier. They get free content (from tax-funded scientists), cheap editing (through free peer review), and sell subscriptions for huge sums (to tax-funded libraries). They have a monopoly and, worse, they are taking money from a notoriously unfrugal consumer: government entities that don't bid.
Publishing, like research, is a public good. It can and should be run by the state. Universities and national labs already have small-run publications which they give away for free. I would bet that, even for these government run institutions, it would be more cost-effective to self-publish. They could put those parts of academic publishing that cost the most (typesetting, printing, mailing) up for bid.
The only tricky point is getting scientists to publish to the new journals--prestige of the journal is quite important.
Anjuta is also nice.
I'm sure Sun would be willing to address each of these points if the government had bought into Star Office (which uses much of the OO.o codebase). There are also independent support providers which would write out a contract & provide training for OO.o.
Word Perfect has styles too!
Reveal codes is an absolutely wonderful feature for fixing broken documents. Not everyone uses Word styles (I'm tempted to say a minority do) & you WILL get broken, kludgy documents. If for no other reason than this, it would be nice to see where codes start/stop.
It is nice to see exactly where an image is anchored or when a hyphen/spacing is breaking/nonreaking and when these or line/page breaks are optional or forced.
It is also extremely useful to see when a STYLE starts/stops! Third-parties sell an atrocious hack to put a reveal codes feature into Word. The real thing is better.
It is the next best feature to using transparent plaintext formats like docbook/LaTeX, where you can get the same info.
While acroread 5 is horrible (and the parent therefore deserves the Funny mod), the beta version of acroread 7 is a nice enough GTK app. I still have complaints & it isn't enough to switch me off of xpdf, I no longer cringe when I need some peculiar features from acroread.
ck is a nice patchset. For something with slightly more, CKO offers everything ck does with Reiser4, Supermount, Alan Cox's -ac patchset, software suspend, updates to libata/ALSA/Bttv, and more.
If anyone knows how to donate small amounts of money to the developers, please let me know: both ck and cko are on my list of projects to eventually donate to (linked to from my URL).
If you want to benefit from it without learning it, you can use a number of GUIs. Scientific Workplace on win32 (commercial, but good to push on those using Word) or LyX (F/OSS) for nearly any platform or many others. Even abiword can write LaTeX!
It isn't difficult to learn & becomes much more powerful when you eventually ditch the GUI & either use a quality TeX-focused editor like KILE (KDE), TeXnicCenter (win32), TeXShop (OS X) (all F/OSS) or your favorite multi-purpose editor. I prefer vim with LaTeX-Suite.
The best way to learn is to look at other code. Either get some from peers, from the net, or make some in either the GUIs or the friendlier editors. Then just write.
If you need a reference, you can usually learn to google for how to do something (or post to comp.text.tex). I maintain a list of www links. You might find something useful, but I can't suggest the best starting point from that list. The best introductory book I've used is Guide to LaTeX. The other books in LaTeX Companions are also excellent for reference, particularly The LaTeX Companion.
Nearly all the programs I use are F/OSS, so I use F/OSS operating systems. They just seem to work better. Even if I had a Mac, I'd probably need to use a PowerPC version of Linux.