14

According to John Fultz

[...] if you want to influence WRI's priorities, the way to do so is to find other folks who have similar concerns and make yourself heard through the Support team. [...]

Which is not common knowledge.

So posting a topic tagged and saying it was already reported actually decreases the chance of it being fixed or moving up in Wolfram's bug-fix priority queue.

I think this is so because someone, who is not aware that a serious bug isn't taken seriously by Wolfram Research unless they get several reports of it, won't report it after seeing that it is already known.

How can we overcome that effect while keeping Mathematica.SE's bugs repository clean and up to date?

  • 5
    I strongly suspect that Wolfram Research takes other things into to consideration when assigning the priority for fixing a bug. I believe what John Fultz wrote about is only one of many criteria that influences WRI. It might be true that it is the only criterion which user's reports can influence. – m_goldberg Sep 17 '16 at 13:50
  • @m_goldberg well. of course. I don't claim the e-mail traffic is a deciding factor but it affects the queue the way I've described. – Kuba♦ Sep 17 '16 at 16:45
  • 2
    While additional weight may indeed sometimes be given to bugs reported multiple times, serious bugs are taken seriously, whether they are independently reported by 1 or 10 people. – ilian Sep 19 '16 at 4:43
  • 1
    @ilian I appreciate developers' work, support and so on, I just don't agree with how resources are allocated (away from FE) or with long lasting bugs. I know it is business and choices have to be made. I just want to emphasize, maybe not critical, but fundamental problems. – Kuba♦ Sep 19 '16 at 11:29
8

Here is what I propose, an example of a question template that should be used when you create a topic about a "feature" you find serious:

Let's say it was already confirmed by WRI Support:

(standard header)

Bug introduced in Vxxx or earlier, persisting through Vxxxx (or similar according to relevant topic on meta)

(addition for serious/annoying bugs that weren't fixed yet)

If you find this bug serious/annoying please report it to WRI Support as it will move it higher in a priority queue.

(if the question body is compact enough one can use as a report:)

You can use this question body as a report content.

(or alternatively)

At the bottom of this post there is an original report I've sent to the support, feel free to use it.

(relevant part of conversation with WRI Support)

A support case with the identification [CASE:xxxxxxx] was created.

[...] I was able to reproduce the issue [...] I am checking with the development team to see if we have a workaround for this.

(*

question

body

*)

  • 8
    I would prefer to not turn this community into a WRI triage board. At the end of the day, WRI is a business and they'll do what's good for their bottom line, regardless of how many people ask them for something or how many places you mention something. If they're interested in listening to the community, they will do so at their own cadence. ¯_(ツ)_/¯ – rm -rf♦ Sep 15 '16 at 14:20
  • 3
    @R.M. isn't it already? But I think I understand your concern, maybe a single sentence then "If you find this bug serious/annoying please report it to WRI Support as it will move it higher in a priority queue."? Not a big change to what we already do> – Kuba♦ Sep 15 '16 at 14:25
  • 5
    Maybe we should add something like this to our SE tools palette — users can click "Report bug in <mma.se question id> to WRI" and then we can use Mathematica to parse the question, craft an email body, include the question as a link and automatically send the email to support@wolfram.com? :D If more emails from multiple people is what John Fultz wants, then we should do our best to remove as many barriers as possible (but without cluttering the site) :P – rm -rf♦ Sep 16 '16 at 2:38
  • 1
    Really everybody with a premier service subscription should call ( +1-217-398-6500) Wolfram support and tell them how much we would love the specific bug xyz to get fixed soon. And that they should send greetings to John Fultz. And Toblerone chocolate if he fixes something. – Rolf Mertig Sep 16 '16 at 12:32
  • 2
    @Rolf I'm on Premier Service, and submitted several bug reports, but never in my life called them directly. How does it work? Do you explain them the bug by reading in your code? Or they pass you over to an expert (maybe a dev) who caringly talks you over the issue and suggests workarounds? Is there any benefit in calling support? – István Zachar Sep 17 '16 at 6:24
  • @R.M. please consider posting an answer. – Kuba♦ Sep 23 '16 at 12:05
  • 2
    @RolfMertig John Fultz is likely getting paid more than me (I'm a post doc in HEP). And, I payed $1k+ for Mathematica. He should send me a Toblerone chocolate for me reporting a bug his team introduced. – QuantumDot Sep 24 '16 at 14:47
  • @QuantumDot But you are close,17%: Uncompress[ "1:eJydUU1LAzEQ3dZP1IJXwUMKvQjapR4EvfmJBZHSFTxPs7PdYDZZk9R2f71Odt2tBxVtDo9kZvLmzbzuRI+TVhAEdp3gXsu4enl4TNr+ukHwJDK01XOHYIy5BI6XUiZrPrZLEDkj1PQabF24STDMcm1cdE7X1Ln8Igzn83lfqBgx7nOdhTzLw2ctEwPZyRgtguHp8clQ8X5oQYIRaCOv7AkXrqL1vcYziTcoocC4ErDfCLhd5AatFVqV/5xmvWqibV+Skhq0n0xbBCNwDo0qh0yQWiaBT3UIriSolwhfZ6g4Rl2KsBwNK0gjS7Rhd0Yrx25VzG7wDaWm5JJG+J7fC262/SB+V9Kqt18qKfdZzVhl9vxa9HLeiqrTVH2a9IWwXXf2ciJPchwFjU5fN9JzNKv7fLT0GTjEmAnoWwf8BRc8BTXF0nMIB6dnp4OVfT2goAJHV5BMwsSyHAoGRs/Ii7+YXVL/bPUhRXqlxS5FYVieFlZw6qKtizW39f/VPa4F/NPhpvE7nQ8xS/K9"] // ReleaseHold – Rolf Mertig Sep 24 '16 at 21:56
  • 1
    @IstvánZachar I did not call in a long time. But I am planning to do so. The human touch might have an effect. Maybe it changes something. I have been reporting bugs since 1990. A lot of them actually got fixed. E.g. I am pretty sure I reported some time early nineties that Series[Gamma[x],{x,0,2}] returned incorrectly SeriesData[x, 0, {1, -EulerGamma}, -1, 1, 1], still present in Version 4.2 (2002), and some time in version 5 it was finally fixed. – Rolf Mertig Sep 24 '16 at 22:04
  • 1
    @RolfMertig Smartypants Rolf! But Im at a public university where I'm paid quite a bit less than at national lab. – QuantumDot Sep 25 '16 at 3:51

You must log in to answer this question.

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