Closed. This question is opinion-based. It is not currently accepting answers.
Want to improve this question? Update the question so it can be answered with facts and citations by editing this post.
Closed 3 years ago.
Improve this question
This is not a question with a precise answer (strictly speaking the answer would be best captured by a poll, but that functionality is not available), but I am genuinely interested in the answer, so I will ask it anyway.
Over the course of your career, how much time have you spent on greenfield development compared with brownfield?
Over the last 10 years I would estimate that I have spent 20% on greenfield and 80% on brownfield. Is this typical?
I think it's typical for professionals who deal with customers to spend more time in brownfield development. The reason is that customers typically aren't willing to throw out their existing software to adopt the "latest and greatest" (green) software.
Developers in research or academics, however, may be more likely to do greenfield development. Start-ups as well.
I think that your ratio 20:80 is representative of many/most developers. As to new development: if you are building software incrementally (Scrum, XP, etc) then one could argue that you spend almost all of your time in brownfield development. Except for the initial iteration/exploratory work, prototyping, even when you are building something new, you are already working on an established code base, refactoring and extending. So how much greenfield development is actually green?
Often the problem doesn't just boil down to brownfield vs greenfield. In some cases there is a valid opportunity for a hybrid greenfield/brownfield approach.
I have written an article called "Classic software mistakes: To Greenfield or Refactor Legacy Code" which discusses this exact subject and outlines a range of possible combinations then evaluates the consequences of each.
http://stepaheadsoftware.blogspot.com.au/2012/09/greenfield-or-refactor-legacy-code-base.html
What may surprise some people is that a non technical attribute, company size, will be a big determinant in the choice of strategy and the likelihood of success of that strategy.
Over the past decade or so, I've always worked on software that was used as the center of my company's business. (Both SaaS and a software product.) And while I've always come into the with an existing system (so brownfield), we've usually put out a ground-up redesign/rewrite (so greenfield.) So, to break to down:
about 60/40 brown/green for the big projects, in number
about 20/80 brown/green for the big projects, in time spent on them
and nearly 0/100 brown green for little side projects
So, that is seems to be the opposite of you. It is the nature of the companies I've sought out, and hence the projects. My software is our company's main product, and that means I work on the same code base for years, usually after having created it from scratch myself/ourselves.
And I like it that way.
Related
Closed. This question is opinion-based. It is not currently accepting answers.
Want to improve this question? Update the question so it can be answered with facts and citations by editing this post.
Closed 8 years ago.
Improve this question
I am a college student keen to improve my Programming skills. I have two pathways to follow:
Contributing to OpenSource Projects
Solving puzzles at codechef.com
Which one should I follow?
A good programmer is one who cares about his or her ACE - Accuracy, Clarity, Effectiveness.
A good programmer cares about the Accuracy of code. The easy part is coding the "happy path" because this is always in the forefront of our minds when we design and write the code. But what about potential the exception paths that exist when presented with unexpected input and edge case behaviours of the chosen implementation provided? Care is shown by taking the time to think through all the code paths, investing time in testing, submitting the code for peer review, and having the willingness to accept other's suggestions and make changes when appropriate.
A good programmer cares about the Clarity of code. Whether the code is well structured, expressive, adheres to the Open-Closed Principle, the Single Responsibility Principle, the executing machine doesn't care one bit. But, these are all very important to the next programmer, or yourself, who has to read and understand your code at a later date in order to fix bugs, modify behaviours, or add features.
A good programmer cares about the Effectiveness of code. Does it satisfy all the constraints imposed on it? Not only performance and space constraints, but also aspects that make it acceptable to the end user, the demands on the development and testing timelines by your clients, boss, family. Professional software development is not a precise circumscribed task, like "calculate the determinant of an NxN matrix". It has many constraints and demands, and good programmers are mindful of all of these, and will do their best to manage the them, especially when there is not enough time to satisfy all constraints completely.
So! To answer your immediate question, Open Source or codechef, I'd say that being involved in an Open Source project provides much greater opportunities to practice being a good software developer. So go choose an Open Source project that you care about, and ACE it!
Closed. This question is opinion-based. It is not currently accepting answers.
Want to improve this question? Update the question so it can be answered with facts and citations by editing this post.
Closed 8 years ago.
Improve this question
We develop scientific software and I manage a small group of applied scientists who write great code. A lot of our products depend on stable development tools which we've been using for developing a stable code base. Now the issue is, someone from the management visited an open source conference and was too pleased to see a lot of great tools which can be used internally for free in place of the commercial ones we've been using so far. So he suggested to the management to remove costs of buying the tools we've using and shift to the open source ones. Now I do not have anything against the open source movement but through a small experiment I found that my team is spending a lot more time debugging and maintaining stable code bases for those open source tools .
I'm sure a lot of other program manager's have had this problem so far. Would people relate their experiences and let me know of any studies made on this subject ? i want to present a cost benefit analysis to the management by giving some statistical facts not just empirical evidence. I'll be glad to know some case studies thereof.
I think open source is terrific, but I use a commercial IDE (IntelliJ) for Java development, even though there are popular open source alternates Eclipse and NetBeans. In my experience, IntelliJ is the best IDE, hands down, with a measurable impact on my productivity.
I can't say that it's true of all tools, but in this case it is.
I don't believe that either open source or commercial tools can claim the high ground here, because I can cite good and bad examples on both sides. Blanket statements and "me, too" thinking are usually a bad idea.
Statistics will be hard to come by. 86% of all statistics are made up on the spot.
I would expect managers in a company whose products are based on science to be more rational. You're a small firm - talk it through. If it's not possible in your situation, then no one has a chance.
Closed. This question is opinion-based. It is not currently accepting answers.
Want to improve this question? Update the question so it can be answered with facts and citations by editing this post.
Closed 7 years ago.
Improve this question
As a passionate opensource developer, I produced a lot of software during the years. In some cases, this software became obsolete due to the fact that I moved to another project, and the platforms changed. I frankly don't have time to maintain my past projects anymore, and I don't have many chances to find a maintainer, as my projects are generally very sectorial in nature. In this sense, I expect them to die out of natural selection not only out of lack of personal involvement, but also because they are intrinsically less appealing to a larger audience that could spawn a maintainer.
In simple terms, my productivity does not scale up with respect to my past products.
Do you have any suggestions for this issue? Since "bit rotting" is so frequent in the opensource world, as it moves very fast, I guess there's a lot of stuff out there which is plain obsolete. Should I let my software rot as I move forward in my development targets, or take the effort of keeping it (even barely) alive, even if it does not pay off in terms of users, and personal productivity?
CW as it's definitely subjective.
Software like any other entity has a shelf life. If no longer needed then it will (should!) die. Unlike other things it will always (well if media survives) be available forever and as such can be resurrected if needed. It need not stay alive.
I saw this many years ago looking at the source code to some termulator. It was a very well thought out and elegant to look at and interesting for a newbie coder to read. However it was obsolete even then. Felt kind of wrong but it was dead!
Move on and enjoy doing the stuff you want to. Your productivity will be decided by how much you enjoy what you are doing.
How are you measuring productivity? If you fix half a dozen warnings so a codebase compiles comfortably on a new compiler and thereby resurrect an entire app, then your productivity is huge, many times anything you could achieve by writing new code. Of course, it is as dull as anything and feels like a real drag to do this, but maybe you shouldn't be measuring enjoyment and calling it productivity :-)
Of course, if this is open source and there is no money to be made, then whatever you do has zero productivity, unless you count enjoyment as production!
Bottom line: if you aren't being paid, then you can do whatever you feel like.
Closed. This question needs to be more focused. It is not currently accepting answers.
Want to improve this question? Update the question so it focuses on one problem only by editing this post.
Closed 7 years ago.
Improve this question
Background: There are a couple of concerns that are not core business for us. They are essential to our core business, but we have no business writing on our own, in terms of manpower, time, and expertise. I am familiar and very comfortable with some open-source implementations, using closed-source-friendly licenses, that could fill these gaps. Closed-source alternatives I either could not find, or were crap.
I put together an informal proposal to show my boss, including the original licenses for each project for legal review. Being a business owner that knows little about the world of open-source, he was initially hesitant when he realized some of these libraries were. I tried to educate him to the best of my abilities (I'm no open-source warrior myself), but he did bring up some valid questions that, in some cases, I don't feel I answered as well as I could have.
Concerns (worded from my boss's prospective)
How do we know and ensure there is no malicious code in an open-source project? Read and understand every line? At that point we could have just written it ourselves!
Who do we blame when things go wrong? With support licenses and a responsible party, we can get things fixed. And if they fail to come through, well... you know.
How do we establish or measure that an approach or implementation in an open-source project is sound, efficient, or good quality?
What sort of liability do we open ourselves up to, in terms of licensing [granted, this is more a question for lawyers and an issue of RFTL].
Question: How have or would you have addressed these concerns?
How do we know and ensure there is no malicious code in an open-source project? Read and understand every line? At that point we could have just written it ourselves!
Same problem with closed source. Actually worse with closed source. With open source at least you CAN review it yourself, or you can take someone else's word for it. With closed source, taking someone's word for it is your only option.
Who do we blame when things go wrong? With support licenses and a responsible party, we can get things fixed. And if they fail to come through, well... you know.
Probably the biggest issue. This depends on which particular solutions you're using. Some things are backed by a reputable vendor (e.g. Red Hat) whereas others have virtually no support. But that "you know" is critical here: ultimately there is no way to guarantee that someone will fix bugs that you encounter when you are using closed source. At least with open source you can hire a 3rd party consultant to do the job, for the right price, because you have the source.
How do we establish or measure that an approach or implementation in an open-source project is sound, efficient, or good quality?
The same way you would with any other code? I don't have any better answers for this one.
What sort of liability do we open ourselves up to, in terms of licensing [granted, this is more a question for lawyers and an issue of RFTL].
Yep, have a lawyer advise you on this. Every tech business should employ a lawyer anyway. The answer will depend on the specific licenses you're dealing with and what exactly you plan to do with the software you develop.
Closed. This question is opinion-based. It is not currently accepting answers.
Want to improve this question? Update the question so it can be answered with facts and citations by editing this post.
Closed 2 years ago.
Improve this question
How do you estimate a EAI project using Function point?
FP analysis is inappropriate for integration projects of any sort as it presupposes that you can specify the application up-front. Most of the work in any integration project of non-trivial complexity is reverse-engineering the nuances of the environment. Typically the environment will not be exhaustively documented in the sort of cases you would expect to use an EAI system in.
By the time you have actually done this level of reverse engineering to the point of having a complete specification you have done most of the work in the project - the actual development is fairly short and sweet by comparison. Therefore the function point analysis is only providing an estimate for a small part of the system.
As an aside, much of the work I do is data warehouse systems in Commercial insurance companies, where extensive prototyping and reconciliation exercises to produce detailed specification documents are actually quite appropriate to the environment. Typically this takes longer than actually developing the production system as most of the data issues are resolved in the prototyping work. EAI systems have a similar class of implementation issues.
Well given that FP counting is based on storage and end user interface, not sure if its even meaningful for EAI (from what little I remember).
I would say you can't, at least not in a useful way. FP counting is generally viewed as a dubious practice of varying accuracy, doing it to an integration project would just add more fuzzyness.