5

I have this question that I thought was well-suited to a site that describes itself as:

...a question and answer site for professionals, academics, and students working within the systems development life cycle.

When I graduated from college, the phrase "software engineering" had not yet entered academic vernacular and I graduated as an electrical engineer, so I hope you can understand my confusion as to why I thought mine was a sensible question to ask.

Please note that I don't feel bad that it's going to be closed as off-topic. All SE sites need to manage their scope or things get out of hand. However, it seemed obvious (even after revewing your on/off topics page) to ask about how to judge when to exclude browser version support when developing an app on a site that claims to be about the system development life cycle.

After reviewing your various FAQ pages further, it appears this site is principally focused on the internals (how do I plan for my software's end-of-life?) and wants to avoid the externals (how do I protect mystelf from another software package's end-of-life?). Cool.

I ask that you to consider a comment left by @DocBrown and an additional insight from myself for addition to your on/off-topic page. If only to help the old fat guys like me who still haunt the halls of software development. Doc Brown said:

...concerning your "life cycle argument": you are playing with words, the specific life cycle of one specific software product of one specific vendor is definitely not on-topic here.

Therefore, please consider adding to your FAQ page:

  • Questions about the life cycle of specific products are off-topic.

  • Questions relating the the business of using software including (but not limited to) the financial impact of supporting software, when to decide to end support, and how to plan for the end-of-life of dependent applications are off-topic.

Thanks for your patience.

  • Thanks for taking this constructive. Looking at the FAQ page, I think you are right, there should be somewhere a link to this older SO meta question and an explanation that this reasoning applies to SE.SE as well. I see what I can do by myself (I am not a moderator here). Not sure, however, if the FAQ needs to mention that business question are not our scope. – Doc Brown Nov 29 '17 at 20:48
  • ... however, looking twice, I see the FAQ points directly here softwareengineering.meta.stackexchange.com/questions/8197/…, that is already pretty close to what you were asking (and a comment from gnat there linked to the above SO meta question). – Doc Brown Nov 29 '17 at 20:56
  • @DocBrown, the question you indicate is "how do I use product X." I agree that the FAQ is clear about not accepting such questions. The reason I didn't heed it is that I wasn't asking about how to use outdated browsers. I was asking about how to decide when to stop supporting them. Some may see that as a subtle difference, but I did not, because I'm not asking how to use them at all. And thanks for considering this. I'm a fan of following site rules. – JBH Nov 29 '17 at 21:11
  • no, that question is not specificially "how do I use product X". It is "Why can't I ask customer service-related questions here?". But I am not sure that even Microsoft can answer your question: they might have a product strategy for replacing IE by Edge over time, or to implement CSS variables in IE one day, so they might be able to tell you if you ask, but even MS does not has a crystal ball to look into the future how markets will really develop. Nor have we. – Doc Brown Nov 29 '17 at 21:30
  • @DocBrown, Please bear in mind I'm no longer looking for an answer to my question here. I'm simply offering my experience to help newcomers better understand the site. – JBH Nov 29 '17 at 21:43
  • 1
    Understood that, but my point is, after taking your specific question as an example and rereading the discussion about it again, I finally believe we don't need a specific FAQ entry "questions about the life cycle of specific products are off-topic". Questions like yours are off-topic because they ask about future markets, so at best they invite to a broad discussion about "what might come" in the future. And IMHO the help center & FAQ already state clearly that broad, opinionated discussions are not welcome on this site. But that is just my opinion, others might think differently. – Doc Brown Nov 29 '17 at 22:01
  • ... so I guess ""questions about the life cycle of specific products" could actually be on-topic, if they are not focussed on the future, and if they don't require inside vendor knowledge about their marketing and development strategy. Sorry for not getting that completly right when I wrote my initial comment to your original question ;-) – Doc Brown Nov 29 '17 at 22:05
  • @DocBrown, Oh! I see. You're correct, only something along the lines of the second bullet is needed. Cheers. – JBH Nov 29 '17 at 23:30
  • 3
    This isn't that complicated. Evaluating the market share of browsers is clearly not within the province of the Systems Development Life Cycle. – Robert Harvey Dec 5 '17 at 19:25
5

I agree with your sentiment.

Frankly I think we are far too quick to close questions on technicalities. When a generous interpretation could be on topic.

For example, you could read your question interpreting IE as an example of a low user base client. How do we cope with these less valuable sectors of our market during development? should they be planned in from day 1? what about non-failing degradation of the experience? do they require separate product streams? does it affect language/framework choice?

Alternatively it could be a specific technical question about IE support and whether there are technical solutions to supporting older HTML 'standards'.

The danger of being too quick to close questions is that we will end up with only

"according to scrum/kanban/clean coding/pythonicness/REST how should we do X"

questions, and trite answers which just mirror the published documentation; rather than being real life professional knowledge and insight from experience, which generally make far more valuable answers even though they may be somewhat opinion based.

  • 2
    Frankly I think we are far too quick to close questions on technicalities. -- Is there any doubt that the example post the OP provided is clearly and unambiguously off-topic? – Robert Harvey Dec 5 '17 at 19:23
  • @RobertHarvey Yes. It looks like a requirements elicidation question, which is on-topic. – Simon B Dec 6 '17 at 14:34
  • 1
    @SimonB: It's about deciding the importance of browser market share. By that logic, marketing questions are also on-topic. – Robert Harvey Dec 6 '17 at 15:50
  • surely concepts such as 'graceful degredation' are on topic. There is considerable room for interpretation of the question in a broader context – Ewan Dec 6 '17 at 16:05
  • 1
    It's that "considerable room" that is the heart of the problem here. Where do you draw the line? – Robert Harvey Dec 6 '17 at 20:43
  • its not where you draw the line so much as how soon you draw it I think. we should be looking at how to make questions work first before removing them. After all they are the content that drives the site – Ewan Dec 6 '17 at 21:11
  • 1
    attitude like expressed in this answer will likely make SE management feel happy "Spolsky transformed SO to match his vision... It did empower more SO users to ask questions; it's much easier now that there's no need for showing research or understanding of the problem. These users are also a voting force: any meta proposal that tries to do something about the flagging quality of Q+A is quickly voted into the ground. Users want a forum, not a wiki..." – gnat Dec 30 '17 at 8:39
1

I've reviewed your post and the comments here.

I'm going to reopen the post, with some minor edits that help get at things that we can answer. We can't answer the question in the title, since we can't tell you what you need to do to make your project successful. But we can address the points in your body about how to determine what to support.

At a very high level, we define our scope as topics related to the systems development life cycle, and provide a link to the Wikipedia page that defines the SDLC, because we agreed that it was a pretty good definition and aligns with what we think can help explain to other people what our scope is. There are several analysis activities on that page that talk about your very concerns: determining objectives, examining alternative solutions, various types of feasibility studies, requirements definition, design, and others.

Other aspects of our scope may also apply - requirements engineering for how you go about determining what you build (elicitation and analysis techniques), software architecture and design for supporting a broad number of browsers and versions, and release management if you are phasing out support for old browsers.

A good answer to this question would address how you can elicit data and then analyze it to drive various design decisions when maintaining your software.

  • I appreciate your help, but it's worth noting that this is a new development. "Current product" is a bit of a misnomer. I'm trying to determine how much effort to put into how many browser versions. My question was driven by CSS variables, which are not supported by any version of IE and never will. – JBH Dec 7 '17 at 0:30
  • @JBH I think I misread part of the second bullet point in the original question, when you talked about different products. – Thomas Owens Dec 7 '17 at 10:58

You must log in to answer this question.

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