Slashdot Mirror

Take it from ol' Mentifex here -- the hassle isn't worth it.

After years of trying to create artificial intelligence as an independent scholar, one ends up broke, pseudo-alcoholic, unhinged, destitute, scorned and despised by the real programmers, and ready to throw in the towel and let others go on to the glory of achieving True AI.

If you are truly, keenly interested in your quondam independent software project, be a quitter, forget about winning, and just do your project as a bumbling amateur. In freelance AI, for instance, it is just as fun to create the AI Minds as an obscure amateur as it is to run a big independent software project. Why bang your head against the wall for years and years? Just give up and enjoy life. Chase chicks. Engage the services of a Wedding Planner for a mating scheduled around 2028. You have been warned, mon fraire. You could end up in the Ditch like Mentifex. Doom awaits you if you persist in your wild-eyed dreams.

When you realize that you have been wasting your time, your money and your precious youth, figure out some way to set packs of other programmers loose on the same objective that you were foolishly hoping to achieve. For instance, as Mentifex here prepares to drop out of sight into AI amateur status, there is one last-ditch effort to get the consummate pros to take up the glorious task of creating AI Mind Exhibits at educational science museums all over America. Let the museum-goers get wind of your starry-eyed software project goals and let them exude blood, sweat and tears in pursuit of your defunct delusional dream. -Signing off, BTDT (Been There Done That)

Take it from ol' Mentifex here -- the hassle isn't worth it.

After years of trying to create artificial intelligence as an independent scholar, one ends up broke, pseudo-alcoholic, unhinged, destitute, scorned and despised by the real programmers, and ready to throw in the towel and let others go on to the glory of achieving True AI.

If you are truly, keenly interested in your quondam independent software project, be a quitter, forget about winning, and just do your project as a bumbling amateur. In freelance AI, for instance, it is just as fun to create the AI Minds as an obscure amateur as it is to run a big independent software project. Why bang your head against the wall for years and years? Just give up and enjoy life. Chase chicks. Engage the services of a Wedding Planner for a mating scheduled around 2028. You have been warned, mon fraire. You could end up in the Ditch like Mentifex. Doom awaits you if you persist in your wild-eyed dreams.

When you realize that you have been wasting your time, your money and your precious youth, figure out some way to set packs of other programmers loose on the same objective that you were foolishly hoping to achieve. For instance, as Mentifex here prepares to drop out of sight into AI amateur status, there is one last-ditch effort to get the consummate pros to take up the glorious task of creating AI Mind Exhibits at educational science museums all over America. Let the museum-goers get wind of your starry-eyed software project goals and let them exude blood, sweat and tears in pursuit of your defunct delusional dream. -Signing off, BTDT (Been There Done That)

And then what? Your post is one of the many that suggest that he's going to need to pay a programmer. Ideas are cheap, but it takes skills, bla bla...

Well, this is slashdot, many of us are involved in open source projects as a hobby and/or professional applications to earn a living. I'm sure some are even really good programmers.

Yet how many of our incredible projects or ideas are succesful, even once they are functional?

I think this guy needs a lot more than a programmer:

- A good business plan, if he intends to make money at some point.
- If he doesn't really expect this to make money he doesn't need a programmer - he needs a marketing guru that can get a programmer excited about the idea. He also needs to understand that a programmer working for free does whatever the fuck he wants to do. So the OP can forget about 'designing' shit.
- Assuming his stuff needs an internet connection, there are other costs. How's going to pay for the server(s), bandwidth, etc? The free programmer that is already working for free?

Honesty, what the OP needs is not a programmer, is Santa's email address.

PS. While we are at it, here's a damn good open source project that needs a decent marketing guy.

I've no personal experience in the field, but I have heard good things about USB Snoop which is open source.

I would probably have gone with USBspy because I'm not afraid of commercial software, I just prefer the other kind. I'm sure Sourceforge has something to solve the problem but I'm not actively seeking an answer today so it's better if the grandparent does the rest of this work himself.

Ah, and Freecaft did change its name, it's around as Stratagus

• Battle for Wesnoth, a turn-based strategy game with some great single-player campaigns.
• Vega Strike, the game Elite would have been if it had been made for today's hardware (honourable mention to Oolite, which faithfully recreates Elite but with updated graphics).
• Nexuiz, a superb FPS with completely new artwork, levels, and game design based on an incredibly heavily modified version of the Quake 1 engine.
• FreeCol (and, of course, the classic FreeCiv), open source clones of the old Colonisation and Civilisation games, with large numbers of updates (and distressingly good single player AI).
• Blob Wars: Metal Blob Solid, a complex platform game, full of gratuitous blob violence[1]. A sequel, this time in full 3D, was released last month.

With complex and polished open source games in almost every genre being available, it's quite surprising how much people spend on commercial games from companies that treat them like criminals. Wikipedia has a good list - I've not played more than a small fraction of them.

[1] This doesn't quite count as open source. The game is all GPL'd, but a number of images were things the author 'found on the Internet' and are used without a valid license. It was removed from the OpenBSD ports system last week because of this, as the author refuses to address the problem.

although this is a thread about GPU h.264 ENcoding, here is a program that uses the GPU for h.264 DEcoding: http://mpc-hc.sourceforge.net/

Yeah, there is. It's called Audacity ;-)

(I bought a USB turntable and it came with a software CD - Win98 USB drivers, and Audacity for Windows, Mac and Linux! Plus source of course ;-) )

Development is active
https://sourceforge.net/forum/forum.php?forum_id=533966
    GyachI 1.1.48 ghosler 17 2008-09-09

If you have a Yahoo account or aren't opposed to getting one, you could give GYachI a try ... it looks like it hasn't been worked on in almost two years, but video messaging works pretty well http://gyachi.sourceforge.net/

It's slow, crashy and overcomplicated.

Your first two arguments are unprovable flamebait, and the last is a matter of opinion. There are lots of people who think it's fast, stable, and just complicated enough.

It's got an ugly, messy desktop environment and it doesn't come with any decent usable software.

Again, the first is a matter of opinion, and I would think you could at least realize that you're in the minority. Lots of people think the desktop is pretty and well-organized. The last is, again, flamebait. It may not come with as much as your typical Linux distribution, but Safari, Pages, Mail, iTunes, Xcode, DVD Player, and the various iLife apps, among others, are far from unusuable or indecent. And, despite the fact that it doesn't come with as much as your typical Linux distribution, there are many thousands of free and open source programs that you can install.

It's got this weird browser that doesn't render stuff, doesn't have AdBlock and which usually gets replaced with Firefox.

"Doesn't render stuff" is, again, unproveable flamebait. Safari does just fine in rendering tests. You're also showing off your ignorance, as it does have AdBlock. Come on, that's the first link in Google.

It can't play back most videos or music files without expensive shareware.

This is just wrong and uninformed. Those are just examples off the top of my head that I like, there are plenty of other free and open source players out there.

It doesn't even have a usable text editor!

What about TextEdit and Pages is not usable?

If those are too flashy for you, just install vim or emacs. They work fine.

It's utter crap. Ubuntu is already better than Mac OSX. Please don't try to make another crappy OSX Aqua-looky-likey clone thing.

You clearly do not even know what you're talking about. Please spend some time using OS X or at least do a bit of research before you try to troll again.

I use Audiacity on my Mac too, works fine. Links for anyone interested. http://audacity.sourceforge.net/download/mac http://opensourcemac.com/

He mixed Apple iTunes sloppy code with Microsoft Vista sloppy code.

That is why I don't use iTunes or Vista, both have sloppy code in them that cause crashed. When you cross both of them together you crash the system or at least cause it to lock up.

It is also why my G3 iMac was never upgraded to Mac OSX and still runs Mac OS9, because of Apple's sloppy code in OSX. If I convert it to a new OS it will either be Linux or AROS, because both of them are stable and being ported to the PowerPC platform or have a port already.

Apple "borrowed" a lot from Commodore, first it was the Vic-20 Commodore logo key copied as the Apple logo key on the Apple //e, then it was the Commodore Vic-20 and Commodore 64 compact design copied with the Apple //c, then it was the Amiga Workbench and co-processor support for 4096 colors and above with the Commodore Amiga in the Macintosh II (The Macintosh II was basically an Amiga 2000 rip-off after the Mr. Coffee Classic black and white Macintosh series was an epic fail), and then NeXT was an AmigaOS rip-off using BSD Unix (AmigaOS/AmigaDOS was based on the Unix-like TriPOS and Steve Jobs learned from his epic fail to use Unix as it is more like the Amiga to help make Next survive), Pixar ripped off the Newtek Video Toaster that Amigas had used (Steve Jobs saw how Amiga 2000s with the Video Toaster did great desktop video for movies and wanted to borrow that tech for Pixar), and then Mac OSX got the AROS and AmigaOS 3.X look and feel but with the Microsoft Windows bloat. AROS does not have the Windows bloat but still has the AmigaDOS/Workbench "less is more" approach in that it is memory efficient and doesn't need a high end processor with tons of memory to run it.

Basically Apple started to slowly evolve into Microsoft, and Amiga and the Amiga technology evolved into what the Macintosh should have been in 1985, and evolved into what it should be with AROS into modern times.

Apple even is suing people like Microsoft did like Pystar because of its EULA, which is very much like the one Microsoft has. Apple vs. Pystar is very much like Microsoft vs. IBM over OS/2, so Apple is evolving to what Microsoft was during the OS/2 years in the 1990's.

It sounds to me like what you want is K-Meleon. But why not accept that the Web is not a native app, and neither needs nor should have native widgets?

It is also quite easy to customize,both with extensions and full bore rebuilds. Firefox by itself is okay, but with Noscript,Adblock Plus,Forecastfox,and FEBE to ensure backups it makes Firefox a must have in my book,even so far as keeping a copy of Firefox Portable(another nice customization) on my flash for out in the field.

And then if Firefox isn't to your liking there is always Flock,and Kmeleon for older Windows machines(also works with Noscript and Adblock with a little tinkering) and of course there is Songbird which is going for an Open Source iTunes kind of thing and is actually a pretty nice media player IMHO. To me that is what is nice about Gecko,it has enough features built in that a good coder can use it as the basis for all sorts of applications and it is trivial to add functionality through extensions to make the browser YOUR way,instead of what some company thinks is best. This is why despite the buzz around Chrome I'll still be installing Gecko based browsers(Firefox,Seamonkey,or Kmeleon depending on the client/machine) on every machine I service or sell. But as always this is my 02c,YMMV

or k9copy?

That clones dogs.

or k9copy?

I am not as well-versed in this as you seem to be, but from looking at the code, there seems to be both IPv4 and IPv6 in there. IPv6 is enabled by a C preprocessor switch (UIP_CONF_IPV6):

http://contiki.cvs.sourceforge.net/contiki/contiki-2.x/core/net/uip.c?view=markup

Given my general level of paranoia, I recommend overwriting zeros, and five times with a cryptographically secure pseudo-random sequence. Recent developments at the National Institute of Standards and Technology with electron-tunneling microscopes suggest even that might not be enough. Honestly, if your data is sufficiently valuable, assume that it is impossible to erase data complete off magnetic media. Bur or shred the media; it's cheaper to buy media new than to lose your secrets.

Because all data recovery companies have electron-tunneling microscopes on hand for recovery and aren't just running a Linux distro with a modified ext3fs to ignore "deleted" inodes. The longest AES key I've cracked is 28 bits (in Python, no less!). Yet we still use a minimum of 128, more likely 256. It's not the guys running recover I'm worried about. It's the spooks with electron f'ing microscopes and a direct connection to AT&T.

I'm using JFS on my mythtv box, but as a previous commenter said, it's not shrinkable, so it doesn't meet your requirements.

The GParted docs have a list of filesystem features that it can handle. That's probably standard across Linux tools, so it might be a good starting point to narrow things down.

How about the mplayerplug-in for Firefox? Since it uses the mplayer engine, it supports just about any format you can throw at it. You'll need to install mplayer and the codec pack from the mplayer site, of course.

I prefer having the browser launch a standalone player myself.

Since it's not the first result in Google (or the second, or even the first page):

http://xplanet.sourceforge.net/

Indeed, it seems it only makes a static picture, versus being a data exploration tool like Google Earth.

You might check out Vega Strike. I've not spent a lot of time with it yet, but it looks like it's basically a 3D version of Escape Velocity.

because Windows CE used to exist for MIPS based mobile systems. So it can run a MIPS version of Windows CE or MIPS based Windows Mobile.

MIPS is one of the platform targets of AROS which will help turn it into a sub$100 Amiga laptop. I think the MIPS based AROS will have 68K emulation to run the old Amiga 68K programs on it via an emulator.

Also, the graphics had reached a level I consider sufficient to support a good game:
-full 3D engine
-Characters and items were clearly recognizable, not reduced to a crude bunch of pixels like in DOOM due to limited computing resources
-the supported resolutions allowed to check out things at a distance, where earlier games (DOOM again) would make things unrecognizable because they shrinked to a few pixels.

So 2d graphics are insufficient to support a good game? I can think of a few exceptions to that. Even if you want to limit it to RPG/FPS hybrids, check out Ultima Underworld. It predates Wolf3d, and is an excellent game.

Sourceforge's TOS is way worse than that. I recently removed my project, The Gamebook Engine, because of it.

According to Sourceforge's TOS (notice it is a different link from parent's) you have to give Sourceforge the right to do anything they want with anything you publish on their servers. Even if I was willing to do that, which I am not, I wouldn't be allowed to publish things like Qt dlls (which I was) or other people's work in general. Some of the code for my project is scavenged from another GPL project, and the gamebook I converted for the project is a derivative work which I don't have full copyright for. I can't give Sourceforge permission to distribute that.

It is a shame Sourceforge's TOS was not included in the article. But maybe if it was, it would not have been posted to Slashdot.

I'm fairly certain the OP actually did mean links.

Are there any other adblocking proxies I can have a try with? As simple as possible, preferably...

I like bfilter, have you tried it?

Try BFfilter.

From the page:

BFilter is a filtering web proxy. It was originally intended for removing banner ads only, but since then its capabilities have been greatly extended. Unlike most of the similar tools, it doesn't rely on blacklists (although it does support them). The problem with blacklists is that advertisers are always one step ahead. You see an ad slip through, you update your blacklist, and in case it didn't help, you add a new entry yourself. Once I got tired of that, I decided to write a proxy that would detect ads heuristically, much like modern anti-virus software manages to detect many viruses unknown to it.

It works well on most sites, but it can fuck up some sites, and some require custom tweaks. All in all, a decent alternative to browser adblock plugins.

Option A: wine

- use winecfg to set windows version to 98.
- before installation copy the contents of CD1 (allied) to a local directory
- use winecfg to configure a cdrom drive (F: for example) to point to it
- further, in winecfg set the corresponding cdrom label to 'CD1'
- run the installer;
$ wine F:/SETUP95/INSTALL.EXE
- copy the included PATCH.* from the XP_Patch subdir to the REDALERT installation folder
- run the inlcuded update patch;
$ wine C:/WESTWOOD/REDALERT/PATCH.EXE
- run the game;
$ wine C:/WESTWOOD/REDALERT/RA95.EXE

Option B: freera (haven't tried it though)

Happy world domination!

I keep hoping that Google will start releaseing some of their data into the public domain/GPL/Creative Commons.
That Google spy van must be gathering data like speed limits, which streets are one way. Maybe even which are paved and not.

You're right for StreetView (you can still use Google's StreetView data in OpenLayers.org for example), otherwise, Google Maps/Earth licenses data from others (Tele Atlas/NAVTEQ/DigitalGlobe/GeoEye/etc), so they are not the ultimate geodata owner (yet? ;-).

One place missing GPL application is a really good navigation system.

Yes but... do you really need this? When you'll buy your GPS-enabled navigation system (e.g. from Garmin, Magellan, TomTom, etc), you'll be given appropriate software that works with the hardware you just purchased (even the iPhone has (in dev) it's turn by turn nav syst software). You don't "need" to install an open source nav syst.

That said, I agree, a solid open source nav syst would be nice. Roadnav is an example, but I think it's not as mature as commercial offers. The data for such an open source software project already exists on OpenStreetMap.org.

Like most of your comments, you are grossly incorrect.

The Commodore 64 helped get AOL started in 1983 before the Amiga existed, but Apple and Microsoft made a deal with AOL to not support the Commodore machines anymore after that AOL rebate was made.

You are obviously an Amiga and Atari hater that doesn't know jack squat about their history and favor the Mac or Windows PCs even if they were inferior to the Amiga and Atari ST from 1985-1994. Around 1995 Windows 95 sunk both of them as PC Clones got so cheap and so fast that Atari and Commodore could not get the 68K series processors to match what Intel put out. Apple had to convert to the PowerPC to match Intel PCs, then Atari went out of business and Amiga got sold off and later they made AmigaOS 4.0 for PowerPC Amiga systems but it was too late. Even Gateway tried to sell the Amiga, but after Microsoft threatened to take away their Windows OEM license, they dropped it like a hot potato and sold it.

AmigaDOS/AmigaOS lives on as the free and open source AROS and if it had third party driver support and third party business software support like Windows, it would wipe the floor with Windows and Mac OSX because it has a smaller footprint and is ten times faster as the modern Windows and Macintosh operating systems and can run on the same hardware as PCs and Macs. But since Microsoft has the monopoly on what third parties can write drivers for and business software for, they sink alternatives to Windows like AROS, eComStation, BeOS, HaikuOS, Linux, *BSD Unix, etc.

This is not a close source browser that Google is shipping (According to their blogs/information), anyone can fork it and run with what they like/dislike.

It's worth mentioning that this is exactly how Chrome's Webkit engine got invented in the first place. It started out as a revision, then a fork, of KDE's KHTML engine. A lot of us were pretty hard on Apple when it became obvious that they weren't interested in participating in KHTML's ongoing development. But now that they've created a successful, portable, fork that's popular on a number of platforms (including KDE!) you have to admit that they made the right call.

Even so, forks are usually not a good thing. When you decide to fork an OS project, you're opting out of the original community, and basically telling them you don't care for where they're taking the project. It's like getting a divorce. Just as partners shouldn't break up their family the first time they get pissed at each other, it's dumb to pull out of a community just because they don't agree with all your priorities.

This is hard for many software people to understand, since they tend to have big opinions about little things. Which is why the Pidgin IM project got forked in a totally unnecessary squabble over a minor GUI feature that easily could have been made optional. Speaking of which, does anybody actually use the fork?

I worked in a small company that used a software development project life cycle management application called SourceForge Enterprise Edition. You can download and use what they call the SFDE version of it if you have 50 users or less. This is a VmWare appliance where CentOS is the guest OS. We had no trouble with it whatsoever. We looked at what it would cost to upgrade to RHEL and, frankly, that was just out of the company's reach financially.

I have been writing business application software for over 20 years. SFEE is most probably the best life cycle application that I have run across but, ultimately, I was still unsatisfied with it. You have trackers and artifacts that you have to customize in order to get change requests and defects. It's kind of a round peg in a square hole deal. Trackers are too generic, too agnostic. I agree with 37signals that software should be opinionated. That is why I am "scratching an itch" by developing a real collaborative software development project life cycle management application.

I am calling this application Code Roller. This app has a lot of features.

• Users can collaborate on requirements, use-cases, test plans, designs, and diagrams.
• Documents can be attached to any of these kinds of items. Documents are managed with multiple taxonomies.
• All of these things can go through a software development friendly workflow process of review and approval/rejection.
• Time is managed through tasks and events.
• The user can also work his bug list.
• A dashboard style interface shows you at a glance what projects that you are working on and what teams that you are a member of.

For more information, please check out my white papers. I would absolutely be honored if members of the /. community would become beta testers.

You can read all the documentation and code. Open source alternate firmware available for Java http://lejos.sourceforge.net/, C, Lua and many other languages.

No they don't support it. I've had many, many conversations with distributors over the years about this topic. It "works" simply because of the way the tools are constructed. But they provide absolutely no guarantees that your app won't break tomorrow with some update they push, and are completely unwilling to make any such guarantees. In fact it's even possible for you to break peoples systems by distributing software on your own.

Trivial example of how things can go wrong, there's no namespacing in Linux. Let's say I make a game and call it Epiphany, then start distributing it outside the framework of the distributions. What sort of things could happen? Well, somebody else might make a web browser called Epiphany, which then might become a part of the base set of packages. What happens when the user tries to upgrade their distribution? Anything might happen, because you have two packages with the same name (or which both try to provide /usr/bin/epiphany).

In the best case the upgrade will just break and the user will be stuck having to choose one of the two packages. But they can't have both.

In the worst case, I decided not to fuck about with 10 different but somehow identical package management systems and used an autopackage or a Loki Installer. Almost all commercial software for Linux does this sort of thing. Now the package manager will just silently overwrite my game files with the web browser. It won't notify the user it's going to do this - it'll just uncleanly corrupt the game.

So what's the solution? Back when I was involved in distribution of apps for Linux, the usual proposal was to put third party software in /usr/local rather than /usr. Unfortunately no distributor properly supports this prefix, and besides, it just moves the problem around rather than solve it. Sadly there actually isn't a solution for this on UNIX - it's fundamental to the design.

You'll notice that Android doesn't use UNIX style directory trees or package management ... and this is probably one of the reasons why.

The distro no longer exists (try following the URL to the homepage listed).

FAIL.

Development-dead, but still lingering around on sourceforge, apparently.

We used "modules" in a similar situation. http://modules.sourceforge.net/

There was a lot of development where I worked before. Things were written with dependancies to specific versions of programs. They would "probably" work with a later version but the developer didn't want to risk that and neither did the business side, so we implemented modules and let the devoper load the version s/he needed.

"module load perl" would always load the latest version but if you depended on a specific version you could always do a "module load perl/5.5.8" which would set environment variables to only get things from that version.

Worked like a charm.

.haeger

You're thinking of JMRI, and it looks like they're starting to move forward in their case (on the winning side no less).

So there is use for copyright/licensing, though it only looks like the F/OSS groups have it right.

Frankly, given Google's record, I refuse to host any of my projects on Google Code, or to participate in the development of any projects hosted there. I use Sourceforge (has svn and ssh access) and Berlios.

Sad to say it, but Windows Mobile phones are probably the most widely available "open for developers" phones out there.

Is there a good way to compile apps for Windows Mobile on Linux? Or on Windows without buying a copy of Visual Studio Standard or higher? (Visual C++ Express lacks the Windows Mobile SDK.) Is CeGCC arm-mingw32ce any good?

This one looks promising. I might give it a try.

If I can get that working with qt I'd be pretty happy.

gvim + CeGCC + qt => good enough for tinkering with. (if it works)

http://torcs.sourceforge.net/ It runs on Linux, but no network play yet.

You don't need to be a print designer to want more fonts. The list of "safe" fonts that can be expected to work reliably in most web browsers includes:

Arial
Arial Black
Comic Sans MS
Courier New
Georgia
Impact
Times New Roman
Trebuchet MS
Verdana

That's it. NINE fonts for BILLIONS of web sites.

I'm not a print designer. But I make lots of web pages, and damn it, nine fonts is not enough. Typography is the single most powerful and versatile design tool in existence. You can use it to convey emotion, to highlight important bits of a page, to subtly improve reading comprehensibility, and on and on.

Not to mention the specialty uses. Have you ever tried to transliterate Egyptian hieroglyphs on the web? I have, and I had to go the sIFR route to represent characters which are just not available, such as the character shaped like a 3 representing a palatal A sound.

And then there's stuff like medieval transcriptions. How can I post a good transcription of a Middle English romance without the characters thorn, eth, yogh, and wynn? Some of those are available in standard fonts, especially thorn and eth, but yogh and wynn are a lot harder to come by. You can get them using Junicode, but only if your visitor happens to have that particular font installed, which 99.99999% of people do not. sIFR isn't really a solution in that case, because you only need four damn characters, repeated at intervals throughout a fairly lengthy text.

But hey, 640K should be enough for anyone!

A Nokia S60 phone (I have an E61) + Nokia S60 Python interpreter have been enough for my personal development needs. The nice thing is, I can develop Python applications on the road only using the phone itself.

Information about the interpreter: http://opensource.nokia.com/projects/pythonfors60/

The latest version is available at the Sourceforge: http://sourceforge.net/projects/pys60

Whatever media you use, you should use par2 or something similar so you (or whoever; leave instructions) will be able to repair any damaged files when the capsule is opened.

Also, redundant copies wouldn't hurt, either.

I swear that I remember a Slashdot article about a year ago that was precisely this: Printing data in a compressed format, which could later be scanned in and converted. What was found is that there were a lot of cool things that they could do to put a lot of data into a small space: Different shapes (square, circle, triangle, circle with a dot in the center) and different colors, mainly. It was very cool stuff and at the time they were able to store quite a lot of data on a single sheet of paper.

I'll leave the exercise of searching for the article up to you, but I did run into this on SourceForge the other day:

http://arss.sourceforge.net/

It turns sound waves into picutres and back again. Similar concept, very fun stuff.

DenyHosts is a handy little script that watches your ssh port, looking for brute force/dictionary attack attempts. Then it blacklists those IP addresses. You can also set it up to share your blacklist with others, and/or to update your own blacklist with what other users have found.

I've been using a package called DenyHosts for about 2 months now. It's in the Debian repos. It just reads the auth.log file and blocks ssh login attempts based on the parameters that you set. It's cut back on my login attempts by about 40% since I started playing with it. It helps a great deal even if you are doing password-less logins, because it will block based on the user, whether it is valid or not, root login attempt, et al. denyhosts.sourceforge.net It's worth looking into as an extra layer of security.