15

At the last Mma Technology Conference in Frankfurt, during the Q&A session I rised a question on the Wolfram plans to enhance the Mma capabilities in analytic calculations. There had been a short discussion and Conrad Wolfram gave an answer. In short his answer has been that Wolfram Research awaits an input from the community with explanation of what additional functionality the community regards as needed.

I would like to generate this discussion at the Stack Exchange. I have my personal answer to this question which might serve a nucleation point for a collective answer.

The question I am asking in the present post is: where is it the proper place to start such a discussion: here or on Stack Exchange Mathematica? Elsewhere?

Please keep in mind that it is in our common interest that the discussion should be as much visible as possible.

  • 1
    This has been brought up before, if memory serves. The consensus I remember was that petitions for new Mathematica features are not kosher for SE. – J. M. is away Jun 15 '15 at 12:48
  • @ Guess who it is. I also remember several such discussions in different places. In this case it is not a petition. It is just a response to the expectation of Conrad Wolfram. – Alexei Boulbitch Jun 15 '15 at 14:26
  • 2
    I've recently watched the John Fultz (Wolfram UI Tech.) talk from Technology Conference 2014 and he is "seriously" willing to consider any good ideas from the community, clearly mentioning also SE. Here is the link (wolfram.com/broadcast/video.php?v=1246) : see the part "Some things on my radar" (40'21") and in particular "Your ideas" (44'56"). – SquareOne Jun 16 '15 at 9:36
  • In some sense questions such as mathematica.stackexchange.com/q/6169/1089 are feature requests which attract attention from the community, if not from WRI? – chris Jun 18 '15 at 7:01
  • 1
    Could the site have a "Feature request" tag that OPs or perhaps monitors could attached to a question to make it easy for Wolfram to identify such things? Not so different than the "bugs" tag. – Jagra Jun 23 '15 at 15:06
  • @Jagra following your suggestion, I have created such tag: one has to advertise and use it I guess now :-) – chris Jun 29 '15 at 8:27
  • @chris -- Great. Thx. – Jagra Jun 29 '15 at 12:59
8

There's an (old) discussion on the topic on Wolfram Community. I think that's a better place than SE since it's more discussion oriented.

  • 5
    Thank you, may be you are right. Your words that it is old suggests that it is also useless, are they? – Alexei Boulbitch Jun 15 '15 at 14:27
  • 3
    @Alexei I think the big problem with these threads is that usually many people post esoteric suggestions that are not of general interest, or make suggestions for functionality that they could easily implement themselves if they were committed enough to wanting it, or that are too vague to be actionable. After too many such posts the whole exercise becomes a disorganized mess and nobody will take it seriously, least of all WRI. There are some very good suggestions in the linked thread, but also some pointless or trivial ones. – Oleksandr R. Jun 16 '15 at 11:38
  • 3
    Perhaps another possibility can be to highlight bugs and limitations in existing functionality that we think need to be fixed: for example, making FullGraphics and AbsoluteOptions work again after version 6, or the promotion of specific useful undocumented functions to the top level. This may result in a more focused effort that WRI will pay more attention to. – Oleksandr R. Jun 16 '15 at 11:39
  • 1
    @OleksandrR. yes, I agree very much. I am quite frustrated that my sillyness about Union is still on top there. I talked about it with Vitaly once, but he didn't delete it. Anyway while I think discussions about this are very valuable, a nice and short compilation of our favourite requests on this site could be a very nice addition. I suppose one of us should add an answer here. – Jacob Akkerboom Jun 16 '15 at 13:38
  • 1
    @AlexeiBoulbitch I'm really not sure if decision makers from WRI ever looked at the thread. I guess some have. In its current state it really is quite long and has many not so reasonable suggestions, so it might indeed be useless at this point. I was still tempted the other day to post, but then I didn't. What I want the most is still bugfixes bugfixes bugfixes!! – Szabolcs Jun 17 '15 at 9:49
  • 1
    @Szabolcs I understand your attitude. The difference between our points may, however, be engendered by the difference of the foci of our interests. Mine lies in analytical calculations. Hence I suffocate by the inadequate analytical instruments of MMa,and must make lots of programming to achieve my goals. This, however, distracts me from the aim of getting the results to the tool of writing code programming steps leading to this result. Further, at mani occasions Conrad Wolfram mentioned an idea to turn the Math education into a massive using Mma. – Alexei Boulbitch Jun 18 '15 at 8:25
  • @Szabolcs There are different aspects of it which I do not discuss now, but there is one within the present scope. It is that the school children can presently easily do (some) mathematical operations that Mma is presently unable to do. I am speaking about transformations made in parts of complex expressions. That is, one can program such things, but it requires some programming knowledge and skills, which are not granted in the case of the school children taken in the mass. It also requires some additional time, while making them by hand does not. – Alexei Boulbitch Jun 18 '15 at 8:32
  • @Szabolcs I conclude that there is a gap between things declared - at least, by Conrad Wolfram (and very attractive from my point of view too) - and the real capability of Mma at present. – Alexei Boulbitch Jun 18 '15 at 8:38
  • @Oleksandr R. Yes, you are right, I agree. Even more I have already participated in one (or a few) such discussion. My point was the answer of Conrad Wolfram that they await our input. That might have been just a matter of politeness, of course. Very likely. But what, if not? – Alexei Boulbitch Jun 18 '15 at 8:43
  • @AlexeiBoulbitch in that case I would expect that WRI would give more guidance to interested parties about how to approach this in the most productive way. I am sure that they do not actively disregard user suggestions, and that they will pay attention to them if they are good enough and submitted in an appropriate manner. But, they surely have more than enough work to do even without our input, so I suspect getting a suggestion implemented is more a matter of being in the right place at the right time, and talking to the right person. – Oleksandr R. Jun 21 '15 at 2:07
  • 1
    @AlexeiBoulbitch This already happens not infrequently on this site, but it is limited in the sense that most WRI employees (quite reasonably) probably do not want to spend their free time talking about their work with customers of their company. The ones who are passionate about the product, generous with their time, and engage with the user community are a great resource, but we cannot expect WRI as a whole to work that way. If someone makes a well thought out suggestion that someone at WRI likes, it will probably receive attention, but I think we don't have much control over this process. – Oleksandr R. Jun 21 '15 at 2:12
  • 2
    @AlexeiBoulbitch their primary consideration for new features is probably whether it will help sell more copies of Mathematica. For this they have to either outdo a competitor or move into a new market. For example, I am not convinced that the financial functionality is all that worthwhile, but it opens up a new market for them. We can all agree that FEM is an incredible new feature that enables many new applications, but the primary motivations for implementing it were probably (a) because Stephen Wolfram wanted it and (b) to try to encroach on COMSOL. – Oleksandr R. Jun 21 '15 at 2:32
  • @Oleksandr R. I agree 100%. It is also how I see the present situation. There are, however, Maple and few other analytic softs, some of them being open sourced, who might overcome Mathematica in the analytic things, if the situation will not be improved. As it stays now, WRI disregards that. It might be a general belief that the time for analytic calculation is over, and the related market, therefore, shrinks. Are they right? If I look at the fraction of analytic questions on the stack exchange, it is small. So they may be right. This is one side of the story. – Alexei Boulbitch Jun 22 '15 at 7:44
  • @Oleksandr R. The other side is my personal conviction that the computer numerics, though presently a great thing, is a fashion, that will become much less acute as soon as we come closer to computational limits. Not now, and may be, not soon, but it will happen. Then the analytics will be revitalized. This is a very future music, however, if any. – Alexei Boulbitch Jun 22 '15 at 7:52

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .