Marcus wrote:
nnk wrote:
Mike Novack wrote:
If you want to use a program as an opponent to learn from:
a) You want one that is significantly stronger than you are so that you will get punished when you make a mistake (a human player your own strength might let you get away with bad habits). But the game is hopelessly distorted at handicaps > nine stones. In other words, you want the program to be playing in the range 9-3 stones stronger than yourself and then you take the appropriate handicap or a triffle less than adequate handicap.
Note: What you will be learning this way is "how to make use of stones already placed" but that is a rather significant part of what go is all about.
b) I don't know why people are so keen on "free" (as in "fee lunch" --- the orginal "free software" concept meant non-restictive licenses, not that the the software would be "no cost*"). Yes, it's always nice not to have to pay for something but nothing wrong with the more usual situation. Software engineers need to eat too! Sorry, in my experience only a very tiny fraction of the people who object to being asked to pay for software are in the habit of providing their own services to others gratis. In other words, it's not wrong/evil when somebody decides "I'll create this software for you but only if you pay me for it". You then get to decide whether you want it badly enough to pay for it.
* The original rallying cry was "software for the price of a book"
by truly free i mean open source, not free as in beer. i am generally not interested and avoiding "free" (non open source, but with 0 pricetag) software, the only exception in the past few years is the kgs client, and even that makes me unconfortable (for instance, i don't buy computers with nvidia video cards because i would be forced to use non-open source software to get them to work properly, this often costs me money, so i don't mind paying for software, even though indirectly in this case). in my experience, the vast majority of people who prefer commercial software (because "it is better") will not pay for it, my personal view of things is: if it's that good, pay whatever the price is, if not, find something which suits you better, either lower price, or free/open, but do not "steal". in my experience, the vast majority of "open source zealots" will share that view, and stand by it. so please don't pick up the stone before you are sure which way you are going to swing it.
further more, there are several examples already of companies making money from open source, so maybe things are not as black and white as some may think. in general, open source (and multi-platform) is good for everybody (it's obvious, just look around you). i won't go into details, but these days, being involved in open source projects is actually a very good selling point for any coder, some will almost require it.
the original cry was "free as in speech" (subtitle: "we're coders and we want to work together, you greedy bastards"

)
the original (as well as the current version) of the license is not non-restrictive as such, it is restrictive enough to protect the freedom of the code (some will say it's "viral"; maybe, i hope it works

). i am talking gpl here not bsd or the likes.
anyway, back to the thread: i still think gnugo is good (strong) enough for a beginner (i know it can kick the s%$^ out of me), but still i would not recommend it except in small doses: play humans, and have fun! i would personally not recommend going anywhere beyond that "the interactive way to go" link before playing at least 100 games or so against humans. that's just my humble opinion, of course.
On Open Source: Yes, I pay for my commercial software, and yes, the general trend is that the commercial software fits my needs FAR better than the open source alternatives. That isn't to say that I don't use Open Source. I do have several Open Source products that DO fit my needs in small ways. I think the difference between my own attitude and the attitude of my colleagues who are much more into Open Source is that I'd rather NOT spend my time on the computer, given the option. So, I spend money to make sure I spend as little time on the computer as possible. (Heh, odd attitude for a programmer/support technician, I know). Sometimes there is no software that fits my needs, so I have built my own tools in the past. In general, though, those tools are hacky and very specific (throw-away code) so it is never released publicly.
glad to hear that, you are, unfortunately, a minority (sorry, you probably know it already), in the sense that you are actually responsibly paying for what you like. my hat goes off to you, but one note: i didn't mean you when i said most won't pay, so, unfortunately, my statement still stands (it was obvious from your wording that you expect one to pay anyway, so it would have been stupid for me to target you specifically with my statement

)
i know what you mean, i hate spending too much time on the computer. perhaps it's exactly because it's your job, and you'd like to do something else with your free time than what you already do all day at work

. i know what you mean about "hackish" tools you really only can use yourself, been there done that (though i'm no programmer, so probably at a different level)
Quote:
The right tool for the job, I say.
agreed. the unix way, isn't it?

. i tend to see it on the long run though, and i discovered it's much more efficient for me to spend a bit of time to get things to work the way i want them, chose my software carefully, make sure it's open and "alive"; for instance, i have been using the same "desktop" for almost a decade now, it took me days to slowly get it to work exactly as i wanted it, but i enjoyed the process. ffwd all these years, i can still just get a new computer, install the (current) package, drop my 10y old config there, and i have my desktop as i know it and love it. it's not pretty, it's not something anybody else would like using i guess, but it's perfect for me, and saves a lot of time. had this been commercial stuff, i would have had to re-learn, re-configure and so on several times, supposing it wasn't dropped and discontinued entirely.
but your point is inportant: open source is not "for free", even if only for this: sometimes you need to spend time to get it to work how you want it. in my experience, it's rarely any different with ocmercial stuff, but that's besides the point. "there's no free lunch"

anyway, sorry for the offtopic rambling. i guess we should give it a rest, i think it's quite clear that our opinions are merely complementary anyway

appologies to the original poster. i hope he at least found it entertaining
Quote:
On gnugo and other robots: Playing gnugo is boring, in my opinion, but some people enjoy playing against it. It's a personal preference. For me, Go is a social endeavor, so I much prefer playing against someone.
i somewhat agree. i tend to prefer a more imperfect human than a computer, as an oponnent. imho, this game is about two people playing eachother. gnugo and other "robots" are great though, for certain purposes, and i think the effort put into them and the level they have reached today is impressive. i think it might also soon push the elite of go to "re-invent" the game, and look at more unconventional approaches. i vaguely believe that, unlike chess, the inherent complexity of this game can mean there's still ways we haven't explored, and which will be enough to blow a machine out of the water again, without going to 21x21 boards

. but what do i know?

i do agree with you, as i said: i would definetly not recommend a beginner to stay hidden with his computer and play against bots. i did it, it's horrible, and it can all but kill your will to play. you can play them from time to time, but don't make it a habbit. just my humble view.