24
$\begingroup$

Well, this post has nothing to do with code, but relate to Mathematica. This query is deep in my mind for so long time. But when I ask Wolfram support, I am always told some official words, which is not my expectation. I want to know the real truth.

Actually I have only used Mathematica for three years. But even so, I can feel the tendency that the next version will introduce more bugs and the current ones are hardly fixed. This is why many of my friends stay at version 9.0.1 still. Since I am able to get the tendency, users with more than 10 years' experience must also be. I will just give two examples here.

  • The bug of RelationGraph is reported twice by myself, but this bug persists.
  • The bug of Image-Tool, which is very convenient to do some image-processing sometimes, is also reported twice by myself. But the bug is more serious in version 11.1. These two GIFs demonstrate it completely. If we are in 11.0.1 we will get some error messages but we can still get the mask . If we are in 11.1, we get nothing but that error window.

I don't have patience to report the same bug anymore. I hope this will not be regarded as a complaint, but a request to know why Mathematica has become like this. Why fewer bugs existed in the core language than before, but they surge after version 10.0? Another funny thing I think of is, why Wolfram Research doesn't open a bug list, that can help customers know the fixed situation and prevents them from submitting a repetitive bug report. It's a win-win choice. And as I know, Dr. Wolfram is also a heavy Mathematica user. Why can he bear those bugs? Can any insiders tell me something?

Improve this question
$\endgroup$
15
  • 5
    $\begingroup$ Here is the good comment: "A software company I worked for years ago had a triage policy about bugs. Some were so critical they were fixed immediately. Less critical ones were fixed in the next release. Some were regarded as being sufficiently insignificant that they were never fixed. I don't know Wolfram Research's policy but I suspect it is similar." $\endgroup$
    – Alexey Golyshev
    Apr 1, 2017 at 14:48
  • 8
    $\begingroup$ I voted to close because I think any answers to this question will be entirely opinion-based. $\endgroup$
    – Cassini
    Apr 1, 2017 at 14:50
  • 5
    $\begingroup$ Use new versions for new features. Use point releases to fix bugs. And take bugs seriously. $\endgroup$
    – David Keith
    Apr 1, 2017 at 16:32
  • 5
    $\begingroup$ Could you reframe this and make it viable by calling for an analysis of MSE to maybe produce the facts to the story? E.g. couldnˋt the bug tags and headers here be used to give a bugsadded-bugsfixed chart for the Versions? That would be interesting. Otherwise I think this should be moved to chat. $\endgroup$
    – gwr
    Apr 1, 2017 at 16:49
  • 11
    $\begingroup$ I do agree that this is off topic, regardless of whether or not the OP's frustration is warranted. Perhaps Wolfram Community is a better place for it. Those who are voting to reopen: do you have any reasoning that is better than that you are frustrated with bugs? I am too, but that does not make this a pratical answerable question. It is not a question. It is a complaint. $\endgroup$
    – Szabolcs
    Apr 1, 2017 at 18:10

1 Answer 1

Reset to default
7
$\begingroup$

My experience is different. As new versions come out, I think two things happen:

  1. Bugs in earlier versions generally (but not always) get fixed. In this sense, as time goes on, the number of bugs decreases, and Mma gets better and better with time.

  2. New functions are introduced. And when functions are new, there are always issues and quirks that need to get resolved (which is why Space Shuttles blow up). And in recent years, the number of new functions added has been staggering. In this sense, Mma is also getting better with time, but given so many new functions, it would not be surprising if issues pop up that need to be resolved.

I certainly don't mean to defend bugs - I find them irksome and irritating, especially when my existing code or notebooks don't function correctly. But fixing them can also be very time consuming -- and fixing one thing can cause other things to break. If the bug is in some specialised code, there may also be personnel issues, where one person coded up some area of functionality, leaves the company, and finding someone else to work through the original person's code cannot be fun.

Improve this answer
$\endgroup$
1
  • 7
    $\begingroup$ I think there is an important third case, especially in Mathematica: functions are changed, either generalized or optimized. These cases are the ones where sometimes bugs are introduced which are especially irritating to users... $\endgroup$
    – Albert Retey
    Apr 2, 2017 at 18:47

Not the answer you're looking for? Browse other questions tagged or ask your own question.