Actually I am with you bear - give me functionality over glamor any day. However, we are in the wrong thread! Recheck the original post. We have wandered into a room full of fanboys chatting about how beautiful GoPanda2 is, even though it is only half built. Where's the exit?

_________________
Dave Sigaty
"Short-lived are both the praiser and the praised, and rememberer and the remembered..."
- Marcus Aurelius; Meditations, VIII 21

Ed,

I said I was going to stop, but your writing makes me want to explain.

Let's recap.

GoPanda is essentially a stripped down minimalist client. Good looking or not. Let's avoid that issue for the moment.

GoPandaII is being developed now, it seems. I have no first hand information.

Now, go back in time. Pick ur client. Forget how they look. They ALL have LESS functionality than WinIgc, and tgWin (although tgWin was not as well thought out).

BOTH of these are ancient by software standards.

IF you graph the degree of functionality or range and scope of functions available for clients since that time, the graph would be sloping generally downward. It would not be an upward graph.

So there is no reasonable basis to expect that a new version of an existing client would suddenly be created that utterly reverses the generally downward trend and arrives on the scene surpassing these 15+ year old clients. Is there any?

Especially so, if none of the developers are aware of or have used these older clients.

So, ALL that I said, is to take a good close look at WinIgc and figure out WHY I and others say it is/was so good.

Then I offered to send copies of the software to developers, etc...

I apologize if the scope of this thread was intended to be limited to the graphical aspects of the client. I find it difficult to think that the extent of the discussion regarding this new client is limited to the way it looks aesthetically. But if that is all that this thread was supposed to talk about, then it is my error to jump in here. It did not look that way to me when I read it... so, sorry.

Now as far as visual beauty and computer generated images go, I have some personal experience in that department. In 1981 I worked for CGL,inc. At the time it was the world's foremost computer graphics company, supplying product and animation software to the world. Oh, the guys who left right before I got there went to California to do a startup animation company. Smart fellows they were. They called it PIXAR. Then I did Ray Tracing in 1985 at RPI's CICG. My work was on the cover of 1986 Siggraph show announcement and on book covers and in magazines. It took literally weeks to generate an image that I can do on my laptop in a minute or less now. And the computer center that we used cost millions of dollars and took up a big room with massive air cooling gear, and disc drives that in their entirety were maybe 6Gb, which at the time was incredible. In 1985 I had an Arpanet email address. I think I played Go online using an ascii green monitor with x's and o's, iirc...

None of this means that you or someone who wasn't born yet doesn't have an order of magnitude more ability and talent that I do or did, because even then there were people who did.

Actually, all this means is that I am a fast typist and have an opinion. Ha ha.

_-_-

I don't think that that is what he was saying. His point is that objectively, it is lame to be producing new software that is missing features found in software produced in the middle ages.

_________________
Patience, grasshopper.

I think Ed's point is that it isn't objective at all, it's a subjective value assessment. You are dismssing the viewpoint that "the aesthetics is crucial to want to use a client, whereas functionality (beyond "I can connect and play games") is only desirable". Is that because you think it's impossible to feel this way, or just that you don't understand it? That having something that looks good is the key, and the extra features beyond that are a bonus?

The value judgements of what is good / bad in a client are surely fundamentally subjective things, when as the person using it I make my judgement on whether I "like" a client based entirely on subjective responses to its good and bad bits.

Video game companies have had no problems selling poorly functioning, shallow games with fabulous visuals for some time now.

I think this comes down to the design issue. When someone designs a prodcut (which this is) the end results should be taken in. When you design one thing and then later on want to add more, you usually end up breaking what is done. And this comes down to Panda.... Do they want people to have a client that is flexible and can change their environment or do they want to dictate the way you access their systems.

If they want the end user to have flexibility, then create that. Personally the look of the client gives me a headache. I am colorblind and this layout causes me issues. If I am forced to use it, I will not. how can i change this to make it playable for me??? why are the developers quiet on this discussion???? people had chatted with some long time users on IGS and all this was brought up and nothing was done?????

Quite so, it is a subjective value assessment. But that is only part of the story.
There is NO CHOICE on most servers as to what you get, how it looks and what the software does.
The functionality is limited or fixed, good or bad.

So, is the suggestion that this IGS software should copy the servers that came AFTER, and jettison all
the positives that formerly were available? Not to differentiate except by "look"?

However in the case of IGS, we know what functionality the server supports.

We also know the functionality & features that was previously supported by older client software.

The relationship between visual "eye candy" and playing Go seems rather unclear to me, in fact I don't think there is any. Go was played just fine on ASCII terminals, etc.

Go players (I hope) are not the same audience as are video game enthusiasts. Go is a conceptual game, whereas thus far I have yet to see a video game that is anything resembling a conceptual game. Is there one? So I agree about the video games having some wonderful visuals, and being devoid of content.

So, is the suggestion that this IGS software should copy the servers that came AFTER, and jettison all
the positives that formerly were available? Not to differentiate except by "look"?

Discussion #1

• But they said that they are working on implementing the features of the older clients.
• And the older clients still work with the server.
• Therefore, the new client doesn't really affect you one way or the other.
• Should bearzbear really care about this at all? You should be just as happy as you were before.

Discussion #2

• Perhaps you care that developers are catering to the preferences of people like Ed instead of people like yourself.
• In that case, you should be arguing with the developers, or simply voting nay by posting your dissatisfaction with the new software.
• However, you really shouldn't be trying to berate Ed for his preferences.
• It doesn't improve your chances of getting what you want unless you think that Ed will change his mind.

Discussion #3

• As far as computer software goes, my aesthetic values are probably closer to yours than Ed's.
• I frankly find nice looking stones/boards distracting when they are on a computer screen.
• That said, nice looking stones/boards are so easy to implement. It's not surprising that they were among the first features to be added. It's easy and doesn't really depend on how other features are implemented.
• Do you really find that work flow process unreasonable?

Discussion #4: Since everyone has put in their two cents...

• My ideal go client has simple line-art boards, two default automatch settings as the only way to seek games, a clock, no chat, and nothing else.
• More features are good for some people, but not for others.
• Absence/presence of features shapes behavior on the servers and depending on what you want out of the online gaming experience fewer features might be better.
• Then again, I worship at the church of emacs, so perhaps I am being internally inconsistent.

Why should I care?

Good question.

Why should anyone care about anything?
Why does anyone care about anything?

This is an open forum on the topic of Go.

Presumably the "developers" read about what people are saying, and care about what users
want or need? Or are they just developing something to suit themselves and maybe whomever is
permitting or paying to have this done?

Emacs? Hmmm... still around. Heh.

The older client(s) unfortunately have not been updated to reflect the few but important changes that IGS
has made since these older clients were last revised. They won't be either. So for example, if I want to participate in a review I have to switch to gLgO. Something of a PIA.

I like to chat with others that I know about things, games, and while playing (at times), and in pro games sometimes "chatter" (I don't kibitz - do you know the difference?). Nice to do that from the command line in the game window and have that reflected transparently to the main terminal window.

There are two major types of users: click a button and type a command.

I like to be able to do both, and typing is frequently way more efficient and faster than clicking on stuff and opening pop up windows that obscure what ur looking at, etc... other times, click is fine.

On a tangent, WBaduk has very nice variation capability, that is especially nice to see in pro games... super excellent. IGS has no equivalent feature I am aware of. Oh well.

I guess for those who want only the bare bones client, that's where well written and flexible software is what shines... just make the more complex stuff a simple to enable set of options?

What would be nice if a whole lot of folks chimed in and in essence "urged" the developers to work in this way...

I wish the developers luck and good results, as I would surely like a really great new client for IGS.

I am not sure this matters or not.

But i think the big difference in this conversation is in the how a client should be developed. Should it start pretty with minimal functions and then try to add additional functionality while alienating an existing core group of players? Or should you develop a flexible framework that allows users to add enhancements/features at a later date.

I have been in the IT field for over 30yrs. Working as a Systems Manager, analyst, programmer, Assistant Director of IT, Oracle DBA, systems integration specialist, to currently a Remote DBA supporting 32 state colleges and community colleges with their student info system and portal technologies. I have also taught systems analysis and design, C, C++, Java, Oracle SQL, PHP and other web technologies.

The issue we see here is the same that I have faced with many of my students… the want of bling over functionality. This is a pit that you can fall into and when changes are to be made or bugs are detected, it is difficult to debug.

A while back, I was part of a group of igs players who had a discussion with one of the developers on multiple occasions. We discussed our history of what we used and what we like to see with igs and a client. The people in the channel that were part of this discussion had over 100yrs of igs playing time combined, with some going back to the initial days of igs. We knew/know what igs is capable of doing and what is lacking in the existing clients. The developer appeared to be concerned with our suggestions and I was hopefully with a new client and he was going to bring something back for us to test and review. This is something we had done for past developers of glgo and other clients. We waited and waited, finally when the new client was releases I personally was not happy with the outcome. Nothing that was discussed was there.

IGS is basically a telnet server. You can ssh or telnet to the server and can send commands to it, boards are displayed in ascii and the client just takes those commands and outputs then in a slick manner. At one time this interface/commands was open to the public. Is that still available to the public so that a peson could write their own client instead of the new panda one? Will any changes in protocol be communicated with these developers in the future?

But personally, the issue is not what client is best. The issue is how IGS functions and allows clients to interface. A newer backend would allow better tools to be used and provide a more flexible front.

Come to think of this in Go terms, this clinet is just part of joban right now. What is Pandanet's fuseki for this? Like in Go, proper planning and positioning in joban can lead to a good end game. The university system I work for is in the middle of doing this transition and while painful, the final outcome can be good. But people need to know what the vision and final outcome is going to be.

In an other analogy, I worked for a community college and my job was a systems manager and programmer. They had a student system that everyone loved. The gui was fantastic and highly customizable. The problem was not the front end client, it was the back end… The system used isam files that were initially created when they transitioned from punched cards. the programmers just made it work with a gui front end using a telnet connection and COBOL. The main issue when I arrived was it required 2 programmers just to maintain the data files because of the issues that it had. The data conversion to oracle was a nightmare, but we started with solid planning, implemented things using a flexible infrastructure, people knew the road map and could see what the end results were going to be.

So, the basic question for the developers is… Where is vision for this, is this something where the functionality is being directed to them, and is the igs protocol open for us to use and develop our own clients?

Personally with the main page that they display, I will assume the last question will be no, but it is good to ask.

The IGS Protocol is open and you can implement your own client. You can even install a local IGS-like server against which you can test; bear in mind that your client should be implemented with slight variations in protocol for IGS and IGS-like servers.

For the love of clams, make your client open-source!

vi, man. Oh no, another eternal flame war!

It is open source?

IF so where can this be found?

Afaik, since nngs they have not made the source or even the protocol generally available.
Perhaps on request, subject to approval by Panda?

_-_-

The client protocol that is available online is what was part of NNGS (the igs clone) and is a subset of the existing client protocol. For example, it does not inclue review and the new rating system updates.

While it would be a good place to start and you could telnet in and see what review and the rating system looks like. But I am not sure if this would violate the user access agrement to IGS. The other issue is that if pandanet does not know you are writing a client and changes the protocol with out notification, your client would break.

So the question comes down to.... does pandanet want an open door client access or are they going to require all access via their client. It is strange that when all the high fives were going on how the client looks, there were making comments, but when questions come up they are silent.

What I have been doing for a lot of the new commands is using the new windows Japanese client which still keeps a telnet window and looking what it's doing there. I'm not usually too curious, since it's limited what I use IGS for nowadays anyway.

And I can add a lot to this list...
- movies which are shallow and stupid but have amazing special effects,
- shoes which look great but give you blisters and fall apart in the rain,
- cars that look like a dream but are absolutely impractical,
- everything you see about the ever-changing fashion - which is actually a perfect example of society falling for the form itself and getting ripped off while a small group of people in the business making fortunes by convincing you the expensive jacket you wore last month is out of style and now you need to buy another expensive jacket... one of my hot buttons, lol.
- etc.

World is full of these things. We are being actively trained since early childhood to take form over function - because it is easier to sell stuff to us when this is how we think. Just put some junk in a shiny box and you make a fortune! And if people stop buying, just put the same junk in a different shiny box... From the point of view of the producers and sellers - its a dream to have a society of people who fall for that and are even proud of it most of the time. From the point of the consumers - they end up with piles of shiny boxes they can impress their equally shallow friends with.

There are many reasons for why it is so... the most obvious and the most prevailing ones are greed on one side and ignorance on the other.

To me, the question is: is that all good?
Should we perpetuate and even defend such attitudes, or should we try to condemn them, laugh and them, and generally try to shame people into looking PAST the shiny box?

But to answer your question - yes, I do understand the allure of shiny boxes. Or poorly functioning software with stunning visuals.

_________________
- Bantari
______________________________________________
WARNING: This post might contain Opinions!!

I have tried my hand at writing a client library for IGS in C#. This project was shelved due to Coursera.

Here is what I would advise: download NNGS, any Linux distro., VirtualBox, Telnet (for recent Windows versions) and Wireshark. Play around.

Where the protocol differs, the changes are not dramatic. A little reverse engineering is necessary but it is hardly difficult.

Such a client library would have to be kept up to date and modified when Pandanet change their server. This is why it must be Open Source. Once said client library is built, anybody can use it to develop any client according to their own, personal tastes.

I don't think Pandanet will change their server very much, in the future - there are too many clients in the wild, they risk alienating thousands of users. Most eastern users seem to use ancient clients - the "...enjoy a great game" one, for example.

I have no official confirmation that custom client development is allowed or endorsed by Pandanet but they don't seem to suppress such development and even publish some old documents on their protocol for use by developers.

Is it possible to have it start a different browser other than the default browser? I want to have it in a Chrome (srware iron is a chrome fork) app window, that is without toolbars etc...
At the current state I first close the browser window which is opened by Pandanet2, then double click a link with this command line:

Uuugh why don't you just play online with GoPanda2 and if you want to do advanced stuff, you save the sgf and open those with SmartGo which is an amazing go client, just sayin.... I haven't heard any mention of SmartGo here which seems strange considering it's a premier go application. Scratch that, you can play IGS on Smartgo too, really if you want advanced functionality then that is exactly what you want in a go application. If you want minimalistic with wonderful looking interface, board, and stones, then you want GoPanda2. Really, you just have to decide what you want out of your client, because we have both options at the present time. I've been using Smartgo for the last year and a half. SmartGo Kifu for Ipad is amazing though the Windows client is the customizeable and advanced (not to mention more powerful version), and SmartGo Books is just the greatest thing ever. When it comes to playing online though I prefer a simpler interface and a nice aesthetic. If I want to do advanced stuff later, the pattern, joseki, and fuseki matching etc of SmartGo is a great way to go over your games later on.

It's understandable viewing the interface of GoPanda2 which looks beautiful, that it is minimalistic. This means functionality can be included, but I wouldn't expect things that are "small luxuries", or stuff that is useless to the average user. Similarly, very few designers are going to take pride in creating a beautiful looking app and then let you circumvent those visuals for your own. Sounds are a possibility as that isn't very farfetched. You won't be seeing rows and rows of buttons to press in GoPanda2 and honestly, I think that's a good thing, stylistically it goes against what they are after here. Obviously though, what they choose to include should be extremely well executed upon and streamlined for users. For example, you have a game information box you can bring up... I haven't tried looking for one but are there hotkeys for stuff like that? Being able to press "I" to toggle the game info box etc are great ways to maintain your minimalist look while still providing functionality/info without requiring clicking through menus.

Funny, the 'bug report' feature is not working. When you press OK to send the report nothing happens. Cancel button works just fine )).
Sent the report by e-mail (it was about switching rooms and updating players/games lists not working properly).

Sry, had neglected that forum far too long...

I just stumbled across this thread and want to give a few comments/answers to the stuff mentioned here:

- Pandanet doesn't like people to write their own clients - they rather want everyone to use the official ones. If you still do it, they won't sue you - but not help you either. And while you can find a lot of documentation about the older stuff on the web, all newer IGS features are "secret".

- GoPanda2's primary target are users who want to "just" play Go in an easy way without having to know too much about the server. So it's for the "button clicking" users, not the "command typing users". So we made several design decisions in order to simply things, for example hiding the difference between open and looking, or the difference between chatter and kibitz. And yes, only a subset of the actual server commands are implemented by now. Some because we didn't find the time for it yet (e.g. trail), some because we just don't think they're useful. But GoPanda2 does include a console now, so experienced users can still do pretty much what they want.

- The bug report bug has been fixed So go ahead, report everything new you find with that tool - that will reach us much sooner than this forum does