Great to see this much input. Thanks all. A few comments and answers I can throw in that might help.
>...very complex spider that was constantly crawling the forums tagging things
One the one hand, I see how a crawler that was good at auto-tagging posts would be great. But as a practical matter I'd weigh the benefits against the costs of building and maintaining the rules, and manually cleaning up after the exceptions and mistaggings. On the other hand, even if well-implemented how much really would such a system improve search results more than the vBulletin indexing and search tool already in place?
Example: If we kept the big bucket "All Extensions" forum, I'd feel pretty confident setting loose an auto-tagger on that. Most extensions are pretty discrete, do things that no other extension does, have terms that mostly belong to it and no other. I have no such confidence that an auto-tagger would be as successful on the whole of ArcGIS when it comes to the different dimensions of workflows, functions, users, developers, industry application, solutions, etc.
> Jim already said he wouldn't (or at least be reluctant) to institute an forum crawler auto-tagger.
I'll admit to you all. There are plenty of ideas that on the quick face of it seem good, but then we don't implement after weighing the cost-benefit of the idea, and weighing the improvement against what other things are broken in the process. I hope that answers the simple question of "If autotagging would be good, why don't you just do it?". I could come up with probably dozens of terms we use in GIS that mean different things in different contexts. If our categorical buckets had a nice clean set of discrete terms with little vagueness or overlap, sure. I guess for the moment we're stuck on the thought of "will it make things so much better versus the cost?" or "might it actually make things worse?"
>Do you think Esri should move any and all existing python related discussions (like those currently in Map Automation, etc) to the future Python subforum?
I would suggest "no".
Contrast: When we had a sock-drawer "Extensions" forum, the overwhelming request was to split them. We created 12 forums on these fairly discrete topics. There is reasonably very little overlap between them; little complex weaving that connects some of them. Then I plowed backward through time over 500 threads manually and neatly parsed them out into the new set of separate extensions forums. It only took about a day or so. There were very very few that stumped us as to where to put them.
In the case of Python, that will be nearly impossible to do cleanly in a way that makes sense to most of us. Some threads are cleanly about Python syntax, others are really about some operational topic that uses Python heavily, and some are really about some problem to solve that might include later some post about Python code.
So to support in a way what Goh_Raj is saying, the factors which will make it difficult to split out old threads is exactly the same issue that will make it tough to know where to start threads going forward. If I'm doing map automation or if I'm doing geoprocessing, is my topic really about that, or is it about using Python to do that. And then after I pose the question, depends on how people reply. The thread could really dive into Python headfirst, or it could stop talking about Python altogether.
Again we're back to: "I'm using Python to do some GP task. Which forum do I put it in?"
But then, against Goh_Raj's point we have done this successfully before. We've had very successful forums in the past labeled "VBA" or "Java" or labeled "AML", "Avenue", "ArcObjects", or "MapObjects", or what have you, that have helped folks learn how to code against these APIs and solve problems regardless of the type of operation (GP, symbology, map output, projections, some industry-unique application, etc.)
Example: I'm coding ArcObjects to do some Geocoding. I may get a good answer if I post in the Geocoding forum. I may get a good answer if I post in the ArcObjects forum. I may just go ahead and post in both forums. I'd rather let users decide. But this example supports the creation of a Python forum despite it making things more complicated in some situations. Or maybe we need a forum software that allows you to make a single thread appear in more than one forum if you want. This one doesn't.
In my estimation, starting a new "Python" forum will make some usage of the forums more confusing. But in the end, if what you're there to talk about is Python scripting, regardless of what you're doing with Python, you would have a single place to go that would be watched by others who are also primarily wanting to talk about and help you with Python.
>If Jim were to create a new forum for Python, what is the argument to not consolidate existing Python-related posts from other forums there (other than it being a pain in the :o)?
We'd leave the old ones where they are. Doing so would be trying to judge shades of gray with the question Is it white or black? In the future, we would hope that if a user's question is MORE about Python scripting than GP or whatever else they're doing with Python, hopefully they'll put it in the Python forum. If their question is MORE about GP, and the use of Python is helpful yet incidental to the topic being discussed, hopefully they'll put it into the GP forum. If we go that route, that's the best we can expect.
>But if Jim has a magic wand to do that sort of thing, by all means.
LOL. No. I split the extensions ones manually myself. And that had nice clear lines. This simply doesn't.
>Why not just create a new thread called Python, and see if it grows. Keep the existing GP forum, keep all the existing posts where they currently are at, let the process work itself out. In 6 months, if Python is not used, Kill it, and move the stuff in it to other areas.
Splitting the Python Forum posts back into the forums that cover topics they're using Python for would be easy. Black and white. Going the opposite way, that is, plowing through all of the other forums to determine which ones should be moved to a "Python" forum is gray. But I'd rather not consider "ease of reversal" as a factor as to whether or not we should do it.
***There is and will be no clear-cut answer.
***All of the ideas in this thread will make something easier while making something else more difficult.
***What we decide will be what we think helps the most and breaks the least.
Thank you all again for your input. I expect that not everyone will agree with me. That would be strange if you all did. But I ask you to trust that my intent is to do what you all want. It's your forum. It has to serve askers and answers, and changes need to fix more than they break.
