12

Sometimes people ask questions, such as "how do I do XY in ArcGIS?".

I occasionally happen to know a pretty simple solution in another software (usually QGIS), but do not know if it is right to give an answer, especially if there is no qgis tag in the question.

What is the policy on that, if there is one?


Now that I think of it, it is probably useful - the answer might not be the right one for the original asked, but later on might help other people who are searching on the topic.

| |
  • 11
    Some things are much easier in FME than QGIS/ArcGIS but unless it is tagged FME I do not try and give an answer using it. If can be done in ArcGIS then that is what the Question asker is requesting. Some users are locked down at work and can only use 1 GIS software. Comments can be suggested that this can be done on another platform – Mapperz Jun 21 '17 at 15:26
  • 1
    Yes, this seems reasonable. The information is present for others, but not presented as an answer, which it is not, in the strictest sense. I still think alternatives should be mentioned though - occasionally users are simply unaware that they exist! – miln40 Jun 21 '17 at 15:32
  • 1
    As an aside I think "Doing XY in ArcGIS?" makes for a more concise question title than "how do I do XY in ArcGIS?" I came to that position mainly from meta.stackexchange.com/questions/10647/… – PolyGeo Jun 22 '17 at 22:47
8

I disagree with your added thought - don't add an answer for different software.

Remember this is a focused Question and Answer site, not a discussion forum. If your answer doesn't answer the question it will possibly attract downvotes, even if it is the correct answer for a different question (or the same question using different software).

If you believe there is benefit in a Q&A with different software, instead of answering that question, I recommend posting a new separate question (same question for your different software) with a self-answer, if it doesn't already exist.

| |
  • 1
    Agreed. I still think it is reasonable to mention an alternative in a comment, though. That way it is not prominently displayed, but still available. – miln40 Jun 22 '17 at 7:58
  • 1
    @miln40 comments are designed to help clarify questions and answers. They may not remain on a question or answer for very long. I think the advice in this answer is spot on. – PolyGeo Jun 22 '17 at 12:41
  • 2
    What if the OP prefers/uses one software, but clearly says he/she accepts suggestions with other software? Is it fine, then? Or it isn't? – Andre Silva Jun 22 '17 at 15:36
  • 1
    @AndreSilva those questions are usually put on hold as too broad, as they are not focused questions. We recommend if the user wants an answer for multiple GIS packages that they are asked separately – Midavalo Jun 22 '17 at 15:44
  • 2
    I find this approach slightly counterproductive. What makes a question focused to me is its content and level of details, which will express and action to be executed or a concept to be understood. What are the odds someone is trying to post an answer when OP is interested, to wait somebody else posting the same question using a different software? Or the chances such user will post a self-answer? Of course, such types of answers (like every other post) will be there to be judged by community about its usefulness (might have a greater value in unanswered posts, for example). – Andre Silva Jun 22 '17 at 15:55
  • Agree, when OP explicitly says about a specific software, or when the question is tagged with a software and OP does not mention about other software. – Andre Silva Jun 22 '17 at 15:55
  • 3
    +1 If the question is 'how to do X in Y' then instructions for package Z aren't valid unless a comment 'Is package Z an option for you?' is responded in an affirmative way. Even though there are some really good open source GIS tools in some environments users cannot install software or are simply unwilling to learn a new interface no matter how good or free it is. On the flip side if a OP asks for QGIS help you can't answer for ArcGIS, MapINFO, ENVI etc. unless they are affirmed as an option as they may not have budget to purchase such packages. – Michael Stimson Jun 23 '17 at 5:24
  • 2
    @MichaelMiles-Stimson Even if "a comment 'Is package Z an option for you?' is responded [to] in an affirmative way" I think the maker of that comment should follow up by commenting "if you change your question to ask about Z instead of X, or ask another question about Z, then I will be able to provide an answer". The question should not be changed if it already has answers about the original software. – PolyGeo Jun 23 '17 at 23:59
7

The only time that I think answering using other than the specified software (and I think the software being used should always be specified unless the question is about a GIS principle and truly independent of any software), is when the answer explains/states that the software being asked about cannot do what is being asked first.

I do this when I sometimes answer questions about ArcMap by using ArcGIS Pro.

I think such exceptions should be used only to break the deadlock when a question would be otherwise be unanswerable or answerable only with a one-liner of "No it cannot do that".

| |
  • I think this is reasonable/valid not only when it is not possible to accomplish with software A, but when using B can do a much better job than A. – Andre Silva Jun 22 '17 at 19:15
  • 3
    @AndreSilva I think that would be a slippery slope to encouraging "wars" between advocates of competing products and I think that one of our strengths is that we remain product neutral by restricting ourselves to answering the focused question asked. My exception above is only to break the deadlock when a question would be otherwise unanswerable or answerable only with "No it cannot do that". – PolyGeo Jun 22 '17 at 19:33
  • 2
    Take gis.stackexchange.com/questions/244715/… as a recent example. – AndreJ Jun 24 '17 at 18:58
4

I used to think the same - that if a person asked (for example) "how do I read data from a CSV file with lat/long columns" then anyone should be able to post an answer about how to do it with ArcGIS, QGIS, MapInfo, PostGIS, etc.

That way the site is a better repository for information and there would be fewer duplicates. One question covers all bases. Simple.

But... it doesn't really work out that way. It isn't answering the question and if you're advocating a commercial product that the OP doesn't have, then it would be more frustrating for them than anyway.

At best I would say:

  • Post a comment and ask what software the OP has available, before you answer using a different software (and then only if you think there's a big enough advantage to warrant switching)
  • If you are going to suggest something else straight off the bat, make sure it's a free product, or it's an extension to their current software. If there is a cost, make sure the cost is in proportion to the problem to be solved.
  • I think it's reasonable to suggest another product when OP is trying to code a solution for no reason (eg "I don't know GIS, but I know C++, so I'm writing a program to reproject coordinates - how do I do that?")
  • If a question has been open for a week without answer, then I think it's fair to suggest other software solutions.
| |

You must log in to answer this question.

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