Re: [computer-go] My experience with Linux
I agree one hundred percent. Indeed, using *nix because you don't want to pay for the OS is a moral choice (for me this is not the best reason for using *nix but I know that this is the main reason argued). Otherwise, you will copy Windows as so many, many, many people do. I see it this way, the users that install linux have more money to spent in your programs. ;-) 2008/4/9, Don Dailey <[EMAIL PROTECTED]>: > > > > David Fotland wrote: > > Does Linux have a decent development environment yet? After using > Visual > > studio, it would be a horrible loss of productivity to go back to > > vi/make/gdb. Of course the linux command line tools are great when you > want > > them, but they are available on Windows through cygwin, so by developing > on > > Windows I get the best command line tools and the best IDE. > > > > > > > > Since I sell software, building Linux apps is out of the question, since > > Linux users will insist that I give them my work for free. > > > > That's not true. Over the years I have payed for Linux software. Not > very long ago I bought a chess program for Linux even though a perhaps > slightly stronger open source program was available. They made > versions available for Linux and windows (I'm not sure about Mac's.) > > If you actually believe that DOS users are willing to pay and Linux > users are not, I think you under a misconception. Years ago I > marketed a very strong chess program (for the time) and discovered that > almost every chess enthusiast had a copy, but the number I actually > sold was pathetic. And I personally know Windows users that > routinely rip off software. > > I know this is subjective, but if I had to guess I would guess that the > level of maturity and integrity of Linux users is higher (on average) > than Windows users.At least I have that 1 data point, myself > :-) But I have known many Linux and Windows people and my strong > impression is that many Windows users don't seem to feel guilty about > anything they do, and that Linux users have made their choice at least > partly due to a certain amount of personal integrity.Having said > that, I don't mean to imply that all Windows users are cheats, I know > many of fine character and integrity too. > > I don't have any issue whatsoever with making money by selling software > either. I'm not one of those guys that think this is somehow > immoral. I don't believe most Linux users think this either. > > > - Don > > > > > > > > > > David > > > > > > > > From: [EMAIL PROTECTED] > > [mailto:[EMAIL PROTECTED] On Behalf Of Jim > O'Flaherty, > > Jr. > > Sent: Wednesday, April 09, 2008 8:55 AM > > To: computer-go > > Subject: Re: [computer-go] My experience with Linux > > > > > > > > I'll second both the original poster (his troubles with Linux mirrored > mine) > > and the reply (I was completely enthralled with Ubuntu...WOW!). > > > > Jim > > > > - Original Message > > From: Álvaro Begué <[EMAIL PROTECTED]> > > To: computer-go > > Sent: Wednesday, April 9, 2008 10:18:11 AM > > Subject: Re: [computer-go] My experience with Linux > > > > Get ubuntu (http://www.ubuntu.com/). You can ask them to send you a > > free CD. And you should consider getting a decent Internet connection. > > > > Álvaro. > > > > > > On Wed, Apr 9, 2008 at 10:54 AM, <[EMAIL PROTECTED]> wrote: > > > >> I got excited about the free software sometime ago and bought a copy > of > >> Susie Linux. But the installation always hang up at some point and can > >> > > never > > > >> complete. I had to kiss my $20 goodbye and so much for the Linux. > Recently > >> my job involves embedded Linux. For whatever reason we used the Fedora > >> version 4. It looks like the Windows 3.1. The newest version may be > more > >> modernized, which I don't have tme to fnd out. The Linux operatng > system > >> > > is > > > >> about 600 Mbyte compressed. Since we have a fast internet, it took only > 40 > >> min. to download. After downloading we needed to find a software that > can > >> write ISO format on CDs. I failed to find such a software on the > internet > >> and ended up use the trial version of Nero. Then the Nero I installed > >> highjacked my CD drive and I had to unnstall it later. I also tried > the > >> 64-bit version of Linux and the installation never worked. > >> > >> I begin to consder install Linux on my PC at home. With my internet > >> connection speed, downloading 600 MB is just unrealistic. The other > option > >> is to order CD's. They cost $45 and up and I'm sure this cost will go > up > >> with time. So much for the free software. I keeps asking myself what > will > >> happen if the installation fails. I only have one computer and one > >> > > internet > > > >> connection. > >> > >> Not that I don't trust other people's opinion, but people pitched > other > >> things before which we never hear again. > >> > >> > >> DL > >> > >> > >> > >> > >> Get the MapQuest
Re: [computer-go] My experience with Linux
On 9-apr-08, at 13:11, David Fotland wrote: Does Linux have a decent development environment yet? It probably depends on the language. Java has several excellent development environments that are superior to Visual Studio IMO. And they're portable. I believe Eclipse can be made to work with gcc but I've never tried. Since that would pretty much prohibit making a Windows version I see no value as you pointed out. This is one of the reasons I switched to Java more than 10 years ago and haven't regretted it a single day. OK, I started this OS war tongue-in-cheek and am surprised by the response. I don't expect anyone to change their idea of their favourite OS any more than changing their choice of programming language. Every OS comes with their problems and frustrations and it's true that once you're used to one of them it seems to work pretty well. It also goes in cycles. Windows XP is actually pretty decent and so was NT at its time. The Mac OS was awful until they finally got OS X together. At the moment I think it's hard to argue against the Mac OS being the best overall OS available and Macs are now also pretty competitive in price compared to PCs. Installing either OS X or Windows XP on a Mac is totally painless, maybe because Apple suplies all the drivers. As far as I'm concerned we can rest this subject here :-) Mark ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
[computer-go] A problem with selectivity and AMAF bias
Yesterday I noticed an odd phenomena in Valkyria wich was caused by high selectivity and AMAF. In this position (;GM[1]FF[4]SZ[9]AP[SmartGo:1.4] KM[7.5] ;B[ee];W[de];B[ed];W[df];B[ef];W[dd];B[dg];W[ge];B[dc];W[cc];B[cd];W[bd] ;B[cb];W[ce];B[bc];W[bb];B[cd];W[cg];B[cc];W[bf];B[gd];W[hd];B[hf];W[gc] ;B[gf];W[fd];B[fb];W[gb];B[fc];W[he];B[fe];W[ib];B[dh];W[ch];B[ad];W[be] ;B[di];W[ci];B[ha];W[ga];B[ie];W[id];B[if];W[fa];B[gd];W[hb];B[eg];W[eb] ;B[ec];W[da];B[ca];W[ae];B[ic]) there are only two possible moves 1) capturing the last black stone played or 2) capture the ko. Only capturing the ko wins. In this position valkyria will first search 1) because capturing the last stone is urgent. But the search locks into to that move only because there are a strong bias against move 2) in the AMAF evaluation for Valkyria. I guess what happens with AMAF is that alternative local moves (local relative to the first move in a repeated sequence) will always be biased downwards. This is so because playing the alternative local moves after the first one is played is often inefficient because of a duplication of effort. Then since there is nothing in the AMAF scores that indicate that move 2 is any good it is never searched, since the search is so selective that the bounds will not grow large to compensate for the bias in a reasonable time. I do not expect your programs to have the same problem in this particular position. But the problem could be general and I am curious if you have solutions for it if you do. I did implement a crude solution that works in this position and did not make Valkyria lose rating overnight. I added IF (#TotVisits > 500) AND (#Visits > 0.9*#TotVisits) THEN uctQ := 0.5*uctQ; before computing Q := beta*virQ + (1-beta)*uctQ; the effect of this is simply that as soon as a move has been played more than 90% of the time after at least 500 moves has been played in the position then other moves has to be played. This has to the drawback that the search gets slightly inefficient when it finds forced moves. One reason I have this problem may be that I bias the Q value towards local shapes after it has been computed. I should perhaps put weight on high priority patterns by adjusting the prior value of the AMAF instead. I realized this when I read the latest easy-to-read paper from the MOGO team and I will test that as well. -Magnus ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] CG'2008 paper: Whole-History Ratings
Rémi Coulom wrote: > Don Dailey wrote: >> Hi Rémi, >> >> For a while I have considered overhauling the rating system for >> CGOS. My system is ad-hoc and based on gradually increasing K >> factor based on >> your opponents K in the standard ELO formula. >> I don't know if your idea here is feasible for a computer server, >> because presumably the players are fixed in strength, but in practice I >> think some bots change. Anyway, I'm no expert on this but want to >> find something better than what I'm doing and I have considered using >> some kind of whole history approach (such as running bayeselo after >> every round on every game, which of course is not very scalable :-) >> >> - Don >> >> > > Hi Don, > > Maybe you could consider implementing Glicko. Glicko is described there: > http://math.bu.edu/people/mg/glicko/glicko.doc/glicko.html > It should be better than any intuitive hand-made formula you could > come up with. > > Bayeselo would probably produce better ratings than Glicko. Running > Bayeselo from scratch after every round may be too costly. But it is > possible to make very efficient incremental updates: adding a few > games, and running a couple of iterations of MM should be extremely > fast. This would require keeping bayeselo in memory all the time, with > current game results. Since it cannot be done with the current program > you'd have to use my C++ code and somehow incorporate it into the > server software. This would be complicated, and may use a significant > amount of memory on the server. But computation time would be very > short (less than 0.001 second). > > The algorithm I describe in my paper may be overkill for rating > programs. If you look at table 1, you'll see that even when rating > humans, Bayeselo outperforms Glicko. Since most programs on CGOS are > constant, I believe that Bayeselo would be very difficult to beat. I much prefer bayeselo to Glicko. But it is a simple choice. From my reading of the reference you gave, it doesn't seem like the most appropriate choice for CGOS, it addresses issues that are probably of minor consequence on CGOS. Actually, it's pretty feasible to run bayeselo between rounds at the moment - but it may not remain so forever unless I make some changes like you describe. After all the games are played, there is a significant wait before the next round is scheduled, and I use that time for bookkeeping chores anyway.I thought of using only the last year of data too, but I want this to be "all time", not "all year." Using only the last year has other problems too, a program might have a zillion games today but next month drop to nothing - in other words it would not appear very stable. There is another option too. I can run the rating calculations off-line (or as a separate process.) I hesitate to do this because we are accustomed to seeing a game played, then an updated rating.So if I did this, or if I did updates less frequently such as once per 12 hours or so, I would want to have an indicator of how many unrated games are in the queue for each program. But for a server, I don't really think this is really a very good way to do it. Of course I use bayeselo once a month to build the All Time ratings list. I've been wanting to get into the source code because bayeselo is interactive, and because of this not so easy to use in an automated process. Because of that, I had to wrap it up inside a simple to use script that takes command line arguments including sgf files. Internally, the script converts the sgf to PGN files because that's the easiest way to use it when the sgf files are available. I've been procrastinating dealing with the code directly because I dread wading in to figure it out!I don't currently have any overwhelming need. But I would love to see it accept sgf files and command line arguments and not act like an interactive shell.I think I could probably convert it to this usage as an option (still maintaining current functionality but have a switch to over-ride this. ) The server is written in tcl and it's not difficult incorporating C code, most high level languages have an interface for doing that. But I have been debating a chance to lua for quite some time because it's so efficient. In either case, it's not difficult to incorporate C code, but I'm not as sure about C++. You can also use SWIG which is very good at wrapping up functionality to create packages for high level languages.It's not that hard in either case to build a package (written in C) that is usable by the server. - Don > > Rémi > ___ > computer-go mailing list > computer-go@computer-go.org > http://www.computer-go.org/mailman/listinfo/computer-go/ > ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] My experience with Linux
2008/4/9, David Fotland <[EMAIL PROTECTED]>: > > Since I sell software, building Linux apps is out of the question, since > Linux users will insist that I give them my work for free. MS Windows users also insist that you give your work for free, look at emule. Maybe you are wrong and loosing sales. Andrés ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] CG'2008 paper: Whole-History Ratings
Of course there is no perfect rating system.I'm probably obsessing over nothing, there is no overwhelming deficiency of CGOS over other rating systems, but all of them suffer from the transitivity problem, I don't think any of them address that. Most "weirdness" that you see with CGOS is either imagined, or based on intransivities.There has always been some suspicion that the anchor is not "heavy" enough but it's hard to prove.If this is the case, the solution is to use more than 1 anchor which at times we have done. There is also some evidence that when I changed the starting rating for programs by lowering it, the anchor was not able to quickly compensate for the extra ELO point deflation and we had a recession. Based on this suspicion I changed it again a few months ago to be a less drastic change over the previous starting value. - Don Petr Baudis wrote: > On Wed, Apr 09, 2008 at 03:40:28PM -0700, Christoph Birk wrote: > >> On Wed, 9 Apr 2008, Matthew Woodcraft wrote: >> >>> It might be that most of those games aren't visible to the rating >>> system. >>> >> That might explain why a rating system may have a hard time >> to follow. >> Bad data in ... bad data out :-) >> > > But the point is that bad data is what you have in the real life. :-) > > Petr "Pasky" Baudis > ___ > computer-go mailing list > computer-go@computer-go.org > http://www.computer-go.org/mailman/listinfo/computer-go/ > > ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] My experience with Linux
Petri Pitkanen wrote: > 2008/4/9, Don Dailey <[EMAIL PROTECTED]>: > >> > >> > Since I sell software, building Linux apps is out of the question, since >> > Linux users will insist that I give them my work for free. >> > >> I don't have any issue whatsoever with making money by selling software >> either. I'm not one of those guys that think this is somehow >> immoral. I don't believe most Linux users think this either. >> >> > Some of linux people think so but then again those individuals will > not even steal your product as they use only free software and very > strict on what licenses mean. But SW market on Linux is pretty - > other than professionalk SW likeHW simulators etc.- small so I guess > making only for windows if better option . Unless you develop on > something likee QT which is fairly portable. > > There is no question that if you want to sell software commercially, Windows is the way to go. But it's not because of the mentality of the users, it's because Windows is simply a much larger market.Windows is the toy even your grandmother has, it's a fine OS for women and children (just kidding :-) But my point is that Windows is the newbie OS and of course you will get much greater sales there. (I'm not berating it, I realize that sophisticated smart people use and like Windows too.) I think there is a strong desire these days to not put your eggs in one basket, and it seems like every new programming language (or most of them) has platform compatibility as one of their design goals. In fact that is what java is all about and the so called "byte code interpreter" that most high level languages champion. The reason I do a lot of stuff in tcl is that it is really mature and extremely platform neutral. It's not my favorite high level language, but it's under-rated and better than just ok. But it's the only one I could find (when I was looking) that makes it easy to create executables that run on several platforms without hassle.The executables are really just scripts wrapped up with a runtime, but it is a no-hassle way to deal with clients I have done work for that use windows exclusively without requiring me to develop in Windows. And TK is pretty awesome, you can hack up a good GUI in just a few minutes. The cgos viewer is an example of a script I wrote in a huge hurry at one sitting in one morning.Of course it's not finished or polished, but it works good enough until I get psyched up enough to work on it again. I didn't even know if it worked on Mac's until someone tried it for me (and it did and I wasn't surprised.) - Don ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] My experience with Linux
Daniel Burgos wrote: > I agree one hundred percent. Indeed, using *nix because you don't want to > pay for the OS is a moral choice (for me this is not the best reason for > using *nix but I know that this is the main reason argued). > My reason for using Unix, is that it's just far easier to get useful things done. Windows is not a "tool" like Unix is. Windows is really at it's very heart an OS for consumers, Unix is a tool for researchers, hackers, thinkers. Computers are not what they were originally envisioned to be. In the early days they were envisioned to be real computing machines and not email and web browser toys.However, Unix is still much more about harnessing the power of the underlying computer system and is a much closer match to this AI tool it was designed to be. Having said that, I don't deny the heavy crossover.Every computer is used for just about every possible thing and it's possible to use Windows now in pretty powerful ways too. But I'm talking about the very heart of the OS.You can see in most every design choice that Unix is about getting real work done, and Windows is about selling dumbed down computers to the masses. Almost every good thing about Windows was stolen from Unix or other OS's.The whole web/email connectivity thing was started and done in Unix decades before Windows. MS was very late adapting due to Bill Gates blunder. To see how successful Microsoft has been, you will see that most people THINK all the innovation was a result of Microsoft's efforts. A lot of people didn't know about GUI's until Microsoft made the horrible windows 3.1 (and earlier versions) which were real poor rip-offs of what Macintosh already had.Most people have been hypnotized into believing MS was the start. I remember my first exposure to Xwindows on Unix. It was a real shocker for me and I thought, WOW, this is WAY better than that crappy windows 3.1 thing that constantly crashes. Another tribute to Windows advertisement campaign is that when people think of UNIX, they think "boring command line" even though Unix was WAY AHEAD of windows even in this area.The fact that Unix has a command line that is far superior to Windows command shell is not a weakness, but it is portrayed by the ignorant as such. Of course DOS is another example. Very stable, but extremely unimaginative and limited. Multi-tasking had already been around for years, but DOS kept computing for the masses in the dark ages for an unbelievably long time in modern technology terms. The DOS mentality was, "why would anybody ever need to be able to do more than 1 thing at a time?" and I heard those very words spoken by DOS users when I first got excited about multi-tasking. Perhaps those early experiences have colored my viewpoint, but I have always thought of Windows as the OS for people with no imagination or insight.I was offended when I saw the MAC commercials because even though it confirmed what I always knew, I hate the dishonest emotional and unreasoning way it was presented. I just really hate any propaganda like promotion that does not appeal to logic, but only to emotion. Let's not even talk about this same tactic being used in presidential elections. - Don > Otherwise, you will copy Windows as so many, many, many people do. > > I see it this way, the users that install linux have more money to spent in > your programs. ;-) > > 2008/4/9, Don Dailey <[EMAIL PROTECTED]>: > >> >> David Fotland wrote: >> >>> Does Linux have a decent development environment yet? After using >>> >> Visual >> >>> studio, it would be a horrible loss of productivity to go back to >>> vi/make/gdb. Of course the linux command line tools are great when you >>> >> want >> >>> them, but they are available on Windows through cygwin, so by developing >>> >> on >> >>> Windows I get the best command line tools and the best IDE. >>> >>> >>> >>> Since I sell software, building Linux apps is out of the question, since >>> Linux users will insist that I give them my work for free. >>> >>> >> That's not true. Over the years I have payed for Linux software. Not >> very long ago I bought a chess program for Linux even though a perhaps >> slightly stronger open source program was available. They made >> versions available for Linux and windows (I'm not sure about Mac's.) >> >> If you actually believe that DOS users are willing to pay and Linux >> users are not, I think you under a misconception. Years ago I >> marketed a very strong chess program (for the time) and discovered that >> almost every chess enthusiast had a copy, but the number I actually >> sold was pathetic. And I personally know Windows users that >> routinely rip off software. >> >> I know this is subjective, but if I had to guess I would guess that the >> level of maturity and integrity of Linux users is higher (on average)
Re: [computer-go] My experience with Linux
Ok, I'm going to speak up in defence of Microsoft. (I'm not really that fond of them, and I am thinking of moving to Linux, particularly if Vista is as bad as I have heard.) I became anti-Mac when trying to buy one for my then employer. It seemed like a reasonable deal, until the salesman asked me "what industry are you from?". This totally put me off. I was trying to buy a tool to do a job. When I go to the hardware store to buy a screwdriver, they never ask me what industry I am from. I learned to appreciate MS when we kept having problems with machines crashing on the office LAN (all Windows machines). We had deduced that this somehow involved HP printers, which we had a few of on the LAN. I was following a Compuserve discussion group about HP products, where other users were describing the same problem. HP representatives were saying it was nothing to do with their product. Then an MS representative posted there, claiming that when we installed an HP printer driver, if we selected the default installation, it also overwrote part of the OS (a file called printman.exe, which I think did the scheduling) with a buggy one created by HP. My experiments on the LAN confirmed this, we had a ceremonial bonfire of HP install disks, and that was the end of the crashes. But there was no apology from HP, no admission that their buggy scheduler was the cause of the problems. It was people in MS who had traced the problem and published the answer. Nick -- Nick Wedd[EMAIL PROTECTED] ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] My experience with Linux
The difference (and I'm not defending HP here) is that a print scheduler for your OS shouldn't even be *writable* by the install "wizard" for your printer. Imagine an OS environment where a printer is a completely passive device that accepts requests to print onto paper. Imagine that it doesn't pong every device on the network, or every other printer on the network (remember appletalk?), and that there is no automatic "printer discovery". In my opinion, printer "discovery" should happen well before the time that the printer is installed. There's nothing to discover if you install the printer yourself. I didn't "discover" the printer attached to my PC when I plugged it in, I "discovered" it on the shelf of the store that I bought it from. If i'm a network administrator, I should be pretty much aware of when I plug a networked printer into the local network, and it seems to be a reasonable responsibility of mine to first make the decision about which machines should be able to print from it, and then to take the necessary steps to make that happen. Expecting an entire network of machines to do that job for me is not a particularly smart way to manage your network. s. ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] A problem with selectivity and AMAF bias
Magnus Persson wrote: Yesterday I noticed an odd phenomena in Valkyria wich was caused by high selectivity and AMAF. In this position (;GM[1]FF[4]SZ[9]AP[SmartGo:1.4] KM[7.5] ;B[ee];W[de];B[ed];W[df];B[ef];W[dd];B[dg];W[ge];B[dc];W[cc];B[cd];W[bd] ;B[cb];W[ce];B[bc];W[bb];B[cd];W[cg];B[cc];W[bf];B[gd];W[hd];B[hf];W[gc] ;B[gf];W[fd];B[fb];W[gb];B[fc];W[he];B[fe];W[ib];B[dh];W[ch];B[ad];W[be] ;B[di];W[ci];B[ha];W[ga];B[ie];W[id];B[if];W[fa];B[gd];W[hb];B[eg];W[eb] ;B[ec];W[da];B[ca];W[ae];B[ic]) there are only two possible moves 1) capturing the last black stone played or 2) capture the ko. Only capturing the ko wins. In this position valkyria will first search 1) because capturing the last stone is urgent. But the search locks into to that move only because there are a strong bias against move 2) in the AMAF evaluation for Valkyria. I guess what happens with AMAF is that alternative local moves (local relative to the first move in a repeated sequence) will always be biased downwards. This is so because playing the alternative local moves after the first one is played is often inefficient because of a duplication of effort. Then since there is nothing in the AMAF scores that indicate that move 2 is any good it is never searched, since the search is so selective that the bounds will not grow large to compensate for the bias in a reasonable time. I do not expect your programs to have the same problem in this particular position. But the problem could be general and I am curious if you have solutions for it if you do. I did implement a crude solution that works in this position and did not make Valkyria lose rating overnight. I added IF (#TotVisits > 500) AND (#Visits > 0.9*#TotVisits) THEN uctQ := 0.5*uctQ; before computing Q := beta*virQ + (1-beta)*uctQ; the effect of this is simply that as soon as a move has been played more than 90% of the time after at least 500 moves has been played in the position then other moves has to be played. This has to the drawback that the search gets slightly inefficient when it finds forced moves. One reason I have this problem may be that I bias the Q value towards local shapes after it has been computed. I should perhaps put weight on high priority patterns by adjusting the prior value of the AMAF instead. I realized this when I read the latest easy-to-read paper from the MOGO team and I will test that as well. -Magnus Isn't this fixed by never straying from a move in the tree until it loses and then trying an untried move? Or something like that. It wasn't my idea and I don't remember the details, but it seems like it fixes what you describe. ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
