Project Transference [closed] - project-management

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
I would like to know your experience when you need to take over somebody else's software project - more so when the original software developer has already resigned.

The most success that we've had with that is to "wiki" everything. During the notice period ask the leaving developer to help you document everything in the team/company wiki and see if you can do code reviews with him/her and add comments to the code while doing the reviews that explain sections. Best for the "taking over" developer to write the comments in the code under the supervision of the leaver.

Cases where original devs leaved before handing over the project are always the most interesting: you're stuck with a codebase in an unknown state. What I always find intriguing is how the new devs often do their utmost best to comment on how badly designed the code is: they forget about the constraints the old devs might have been under, the shortcuts they might have been forced to make. The saying is always Old dev == bad dev. What do you people think:
I would even call this out as an official bad practice: bad-mouthing the ones who have been before us.
I try to take as much a pragmatic approach as possible: learn the codebase, wander around a bit. Try to understand the relation between requirements and code, even is there is no clear initial relationship at all. There will always be the "aha moment" when you realise why they did something was done this way or that. If you're still convinced something is implemented the wrong way, do your refactorings if possible. And isolate the pieces of code you cannot change: unit test them by using a mocking framework.
Hail to the maintenance developer.

I once joined a team which has been handed over a pile of steaming crap from outsourcing. The original project - a multimedia content manager based on Java, Struts, Hibernate|Oracle - was well structured (it seems like it was the work of a couple of people, pair programming, wise use of design patterns, some unit testing). Then someone else inherited the project and endlessly copy-pasted features, loosened the business rules, patched, branched until it became a huge spaghetti monster with fine crafted piece of codes like:
List<Stuff> stuff = null;
if (LOG.isDebugEnabled())
{
stuff = findStuff();
LOG.debug("Yeah, I'm a smart guy!");
for (Stuff stu : stuff)
{
LOG.debug("I've got this stuff: " + stu);
}
}
methodThatUsesStuff(stuff);
hidden amongst the other brilliant ingenuity.
I tamed the beast via patient refactoring (extracting methods and classes more of the times), commenting the code from time to time, reorganizing everything till the codebase shrunk by 30%, getting more and more manageable over time.

I had to take over someone else’s code of different degrees of quality on several occasions. Hence the tips:
Make effort to take structured notes of any piece of significant information from minute one: names of stakeholders, business rules, code and document locations etc. It is best to dedicate a fresh spiral notebook, so you could tear pages out if you had to.
Make use of one of the better free indexing and desktop search tools available on the market (Google Desktop Search, MS Windows Search will do). Add all document, e-mail, code locations to it.
Before developing anything do document analysis: find everything you can get you hands on electronically on network and printed out docs, make effort of simply reading it. There is amazingly much of useful information even within unfinished drafts.
Mind map the code, architecture etc as you go.
With lesser documented and maintained systems you inevitably will have moments of despair that are likely to push you into procrastination mode. Especially during your first days or week when amount of new information your mind has to digest is overwhelming. At these times it is nice to have someone to remind you (or just do it yourself) to take it easy, concentrate on important things first and revert to making smaller steps in trying to gain understanding instead of trying to leap forward.
Keep taking notes, making diagrams, drawing rich pictures, mind mapping. It really helps to digest the copious amounts of new information, mostly disorganised.
Hei, good luck!

We actually have a specified set of "Deliverables" that has to be present for us to take over a project.
If we have the chance we try to push in one of our folks within the group developing the project at first. That way we get some firsthand knowledgde before our group takes over the code. (in the line of what #Guy wrote)
That being said, the most important part for me would be:
Some kind og highlevel overview(drawing?) of what the code do.
Easy access to ask questions of the people who actually wrote the code
This for me is alpha omega when taking over code and projects

Related

What kind of specs, documents, analysis do you get from superiors when starting a project? [closed]

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
I currently work in a small business (15-20 employees, 5 programmers) where most projects are custom built CMS and a few web applications products.
Since I started working there, I have worked on many projects, but specifications for each project vary a lot. Sometimes we get a little detail, a Word document telling what the client wants, and what we are suggesting (suggested form fields, a short description of display, etc.). Sometimes almost nothing except "do what you think is the best approach for this project/module/request".
My question to you guys, who might work in different kind of businesses, is: How (huge pile of paper? Word docs? Visios?) and what kind of information do you get from your superiors, managers, teamates when starting a project (plenty of analysis, drawings, etc.)? How much detail do you get on this?
Hope my question is clear enough, thank you.
Specs..that's kind of funny...how about never :(.
Seriously a lot of companies assume specs aren't needed, its absolutely unacceptable but this is how it is in a LOT of companies. They assume a one liner and the programmer knows what the program should do, the inputs / outputs and so on.
Unfortunately in my case I have to actually help write the specs..and Im the programmer :(.
I mostly get a lot of verbal direction and I use a voice recorder to record the conversation and transcribe it when I am done. I write my own specs from my customers' words.
Then, as a good consultant should, I take the writeup back to the customer and verify it, and get a signature and build it, and they live happily every after! (no they dont, they change their mind a 100 times)
It can vary depending on what group the work falls under:
Support request - If the change will take a short period of time and is fixing something broken, there is this group. This could be as simple as, "Add Bob to the list of authorized users for that ancient form" where the form is something written years ago and aside from adding and removing users, it isn't touched for fear of breaking things.
Service Advisory Committee request - Items that are up to a few days are in this group as these are kind of like mini-projects as the request may be to create a new form or portal for a group. This could be upgrading some 3rd party software where we have some customizations that make the upgrade not necessarily a simple thing for Operations to do.
Project - In this case there are usually a few Word documents and/or e-mail threads that help nail down requirements in terms of scope, budget, and time. These can take months though there is something to be said for having a prototype to change rather than creating the initial prototype to tell if requirements are really met or not. Course my current project is over a year old, still has a few more months to the timeline and already has a successor coming after it is done,i.e. there is a Phase II to go after Phase I.
Uber project - These merit their own group of documentation and are the million dollar, multiple company projects that usually try to document everything up front rarely works out well here. Thus, there is some adoptioon of agile for these but there are still some growing pains to go through as how we use agile matures. Think installing a dozen modules of some off-the-shelf software that requires both internal and external developers to customize the suite for our specific needs as the software is supposed to be very robust, flexible and help save lots of time and money on how people otherwise do their jobs generally. Think ERP or CRM for a couple of examples here.
We are a 16-person company that creates and supports customized software for small retail shop owners.
The projects we get fall into three general categories (as related to specs):
"Here, automate this form." A sales person explains that our customer only wants this form to appear where they can fill it out and print it to make it look professional to their customer. Our specs is a single piece of paper that looks something like an order form or report. This is always false; they want pop-up lookups, automatic updating from other sources, and "while you're at it" add-ons that more than double the time. These, we've learned to just live in the moment and let the project take its course. By the time we're done, the program doesn't look anything like their original form.
Small changes. Like a simple e-mail explaining that the background color is stale, or a request to sort a report by a different column. These, we just do as time allows.
Big company integrations, where we're tasked with making our software work with some big outfit like Intuit (QuickBook) or FedEx (shipping rates). These often have well thought out documentation and sample code. We get 100's of pages in word documents or pdfs. The problem with these is when their specs are wrong. We find out about inaccuracies when we try to test or certify our integration. In these instances, we usually take longer in certification than we did to originally develop the processes.
In all cases, the real trouble is when a sales person promises a solution to the customer before even asking a programmer what it would take. As recently as 2 weeks ago, a sales person got into real trouble and had to issue a refund (that person is no longer with the company).
None - at least not from management.
Instead, as a developer (and particularly one leading a software project right now), I'm expected to contact my users/customers/etc and work directly with them to come up with our specifications and requirements. The documentation I do request from my team is only what will be useful to the team. I am lucky in that management rarely requests a document that doesn't make sense or won't provide some use to our project.
I currently have a half-dozen or so specs each 60-80 pages. One of them is 80 pages with no table of contents. Good times.
Our Product Managers and senior engineers prepare three planning docs for our data management software projects.
High-level requirements: 1-to-3 sentence descriptions of hardware/software supported or specific feature for this project. (10-15 pages of Excel-like grids)
Technical details: Engineering implementation of each high-level requirements. Up to a page for each, depending on amount of detail. (30-40 pages of filled-in feature details)
Business agreement: Summary of 1 & 2 with engineering schedule and Product Mgmt's market analysis. Everyone signs off on this. (5 pages analysis, 20 technical)
I haven't seen work flows or other Visio-like details in our specs. The prioritized requirements and schedule prove critical, so we understand when to lop things off to save development and testing time.

When do you refactor code? [closed]

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
Do you do it when you’re in the code doing something else?
When your manager approves it? (Seems this never happens)
I guess some of this depends on the impact of the changes. If I change the code and it affects nothing outside of the class, to me that is low impact.
What does it become a design change? When it effect X object or X projects?
I’m just curious how others teams tackle this...
As part of original development (red/green/refactor)
When suggested by a code reviewer
When we've noticed a design pain-point
When making another change, if the refactoring is low impact, i.e. typically not affecting any other files.
If it affects the public API, I generally like to make the refactoring a single source code commit which doesn't change behaviour (and then build new behaviour into another commit). If it affects other projects too, there needs to be consensus over it and I would want to get permission to change their code to go in the same refactoring commit.
I find I refactor when revisiting code (presumably to add/extend functionality) more than 3 months after it was written.
If it takes me more than 2 minutes to discern what a chunk of code is doing, I'll break it apart to make it more immediately understandable (or just add some more comments.)
as soon as all of the tests run.
I work in a large system, so I only change things I have to. It is easy to have bad side effects to changes.
I will refactor sections of code that are performing poorly, not working properly, or needs new functionality.
I never just decide to fix things, I would never be done. if it works, and no one is asking for changes or complaining about problems, move on. life is too short to fix everything.
I often refactor my code when there is a user requirement change or bug fixes. Then there will be a chance for people to review your changes.
Otherwise, I normally don't touch the workable code even it smells.
We found small refactorings are best done while we were working on a bit of code - do what's required, preferably paired.
For bigger things, we had a Technical Debt section on the wall - if you spotted something and didn't have the time to address it, or it was going to take some discussion to solve, you'd add it to the wall and they would be scheduled for future iterations (or when free time cropped up).
Refactoring while you're already in the code is sometimes easiest, especially if your manager does not support the initiative, but if you only change a small part it will break consistency with surrounding parts. In these cases it's better to be selective and, as you suggested, do things that are low-impact. It may also be helpful to refactor long select/switch statements into functions and delay on refactoring the inner code until sometime later.
At a previous job, I was the manager, so I refactored whenever I wanted. At my current job, I'm an analyst so most of the code is not directly my responsibility. When I do write code, I avoid impacting anything that I'm not writing. I have one project which is entirely under my own control and I refactor any time I learn a better way to do something.
We refactor as often as we can. Having unit tests to ensure that everything works pre- and post- refactoring really helps.
Code review processes often help with this. If I touch some code, it gets reviewed, reviewer asks, "why did you do it this way?", I say, "I had to because of (insert ugliness here)". This is a sign that the code should be refactored right after the review is done.
To look at our company, we have decided that our upcoming application release is mostly dedicated to performance optimizations rather than new functionality. This was something we felt was needed and also was requested by some clients.
Therefore we have spent a lot of time identifying performance bottlenecks in our app and reviewing code and refactoring it to make things run faster.
So in our case we did it because management approved us doing it for this new release, because we showed to them how much performance improvement could be gained.
Refactor when needed:
when you need a better understanding of the code you are working on (pairing often helps here), examples are: renaming, method extraction etc.
when the current design doesn't allow for a 'clean' change: at this point you can actually argue with your manager on a value basis (e.g. what is this new feature worth to the project)
I am always making small refactorings in my code. I know as long as I have my unit tests to verify that everything is still functioning properly afterward, I see no harm in doing it as I go. That way you don't get that vague "needs refactoring" feeling every time you work on it.
Now if it requires a large refactoring, it's best to plan for that and set aside some time.
Seems most other posters are resistant to refacotring mercilessly. Of course this isn't possible if the system you're working on doesn't support this through extensive unit tests. But in general, If I can see an opportunity to make the code tighter without spending more than a few minutes or hours at most, I go for it. If I'm not sure what I should be working on, I look for something to refactor.
I refactor when I'm fixing a bug or adding a feature and the process of refactoring makes the code easier to read and easier to maintain.
Following DRY principles vehemently will often be a trigger for me to refactor.
Insufficiently often, thus building up technical debt.
Sad, but so.
Do as I say, not as the team I work on does.

What can you do to a legacy codebase that will have the greatest impact on improving the quality?

As you work in a legacy codebase what will have the greatest impact over time that will improve the quality of the codebase?
Remove unused code
Remove duplicated code
Add unit tests to improve test coverage where coverage is low
Create consistent formatting across files
Update 3rd party software
Reduce warnings generated by static analysis tools (i.e.Findbugs)
The codebase has been written by many developers with varying levels of expertise over many years, with a lot of areas untested and some untestable without spending a significant time on writing tests.
Read Michael Feather's book "Working effectively with Legacy Code"
This is a GREAT book.
If you don't like that answer, then the best advice I can give would be:
First, stop making new legacy code[1]
[1]: Legacy code = code without unit tests and therefore an unknown
Changing legacy code without an automated test suite in place is dangerous and irresponsible. Without good unit test coverage, you can't possibly know what affect those changes will have. Feathers recommends a "stranglehold" approach where you isolate areas of code you need to change, write some basic tests to verify basic assumptions, make small changes backed by unit tests, and work out from there.
NOTE: I'm not saying you need to stop everything and spend weeks writing tests for everything. Quite the contrary, just test around the areas you need to test and work out from there.
Jimmy Bogard and Ray Houston did an interesting screen cast on a subject very similar to this:
http://www.lostechies.com/blogs/jimmy_bogard/archive/2008/05/06/pablotv-eliminating-static-dependencies-screencast.aspx
I work with a legacy 1M LOC application written and modified by about 50 programmers.
* Remove unused code
Almost useless... just ignore it. You wont get a big Return On Investment (ROI) from that one.
* Remove duplicated code
Actually, when I fix something I always search for duplicate. If I found some I put a generic function or comment all code occurrence for duplication (sometime, the effort for putting a generic function doesn't worth it). The main idea, is that I hate doing the same action more than once. Another reason is because there's always someone (could be me) that forget to check for other occurrence...
* Add unit tests to improve test coverage where coverage is low
Automated unit tests is wonderful... but if you have a big backlog, the task itself is hard to promote unless you have stability issue. Go with the part you are working on and hope that in a few year you have decent coverage.
* Create consistent formatting across files
IMO the difference in formatting is part of the legacy. It give you an hint about who or when the code was written. This can gave you some clue about how to behave in that part of the code. Doing the job of reformatting, isn't fun and it doesn't give any value for your customer.
* Update 3rd party software
Do it only if there's new really nice feature's or the version you have is not supported by the new operating system.
* Reduce warnings generated by static analysis tools
It can worth it. Sometime warning can hide a potential bug.
I'd say 'remove duplicated code' pretty much means you have to pull code out and abstract it so it can be used in multiple places - this, in theory, makes bugs easier to fix because you only have to fix one piece of code, as opposed to many pieces of code, to fix a bug in it.
Add unit tests to improve test coverage. Having good test coverage will allow you to refactor and improve functionality without fear.
There is a good book on this written by the author of CPPUnit, Working Effectively with Legacy Code.
Adding tests to legacy code is certianly more challenging than creating them from scratch. The most useful concept I've taken away from the book is the notion of "seams", which Feathers defines as
"a place where you can alter behavior in your program without editing in that place."
Sometimes its worth refactoring to create seams that will make future testing easier (or possible in the first place.) The google testing blog has several interesting posts on the subject, mostly revolving around the process of Dependency Injection.
I can relate to this question as I currently have in my lap one of 'those' old school codebase. Its not really legacy but its certainly not followed the trend of the years.
I'll tell you the things I would love to fix in it as they bug me every day:
Document the input and output variables
Refactor the variable names so they actually mean something other and some hungarian notation prefix followed by an acronym of three letters with some obscure meaning. CammelCase is the way to go.
I'm scared to death of changing any code as it will affect hundreds of clients that use the software and someone WILL notice even the most obscure side effect. Any repeatable regression tests would be a blessing since there are zero now.
The rest is really peanuts. These are the main problems with a legacy codebase, they really eat up tons of time.
I'd say it largely depends on what you want to do with the legacy code...
If it will indefinitely remain in maintenance mode and it's working fine, doing nothing at all is your best bet. "If it ain't broke, don't fix it."
If it's not working fine, removing the unused code and refactoring the duplicate code will make debugging a lot easier. However, I would only make these changes on the erring code.
If you plan on version 2.0, add unit tests and clean up the code you will bring forward
Good documentation. As someone who has to maintain and extend legacy code, that is the number one problem. It's difficult, if not downright dangerous to change code you don't understand. Even if you're lucky enough to be handed documented code, how sure are you that the documentation is right? That it covers all of the implicit knowledge of the original author? That it speaks to all of the "tricks" and edge cases?
Good documentation is what allows those other than the original author to understand, fix, and extend even bad code. I'll take hacked yet well-documented code that I can understand over perfect yet inscrutable code any day of the week.
The single biggest thing that I've done to the legacy code that I have to work with is to build a real API around it. It's a 1970's style COBOL API that I've built a .NET object model around, so that all the unsafe code is in one place, all of the translation between the API's native data types and .NET data types is in one place, the primary methods return and accept DataSets, and so on.
This was immensely difficult to do right, and there are still some defects in it that I know about. It's not terrifically efficient either, with all the marshalling that goes on. But on the other hand, I can build a DataGridView that round-trips data to a 15-year-old application which persists its data in Btrieve (!) in about half an hour, and it works. When customers come to me with projects, my estimates are in days and weeks rather than months and years.
As a parallel to what Josh Segall said, I would say comment the hell out of it. I've worked on several very large legacy systems that got dumped in my lap, and I found the biggest problem was keeping track of what I already learned about a particular section of code. Once I started placing notes as I go, including "To Do" notes, I stopped re-figuring out what I already figured out. Then I could focus on how those code segments flow and interact.
I would say just leave it alone for the most part. If it's not broken then don't fix it. If it is broken then go ahead and fix and improve the portion of the code that is broken and its immediately surrounding code. You can use the pain of the bug or sorely missing feature to justify the effort and expense of improving that part.
I would not recommend any wholesale kind of rewrite, refactor, reformat, or putting in of unit tests that is not guided by actual business or end-user need.
If you do get the opportunity to fix something, then do it right (the chance of doing it right the first time might have already passed, but since you are touching that part again might as well do it right time around) and this includes all the items you mentioned.
So in summary, there's no single or just a few things that you should do. You should do it all but in small portions and in an opportunistic manner.
Late to the party, but the following may be worth doing where a function/method is used or referenced often:
Local variables often tend to be poorly named in legacy code (often owing to their scope expanding when a method is modified, and not being updated to reflect this). Renaming these in line with their actual purpose can help clarify legacy code.
Even just laying out the method slightly differently can work wonders - for instance, putting all the clauses of an if on one line.
There might be stale/confusing code comments there already. Remove them if they're not needed, or amend them if you absolutely have to. (Of course, I'm not advocating removal of useful comments, just those that are a hindrance.)
These might not have the massive headline impact you're looking for, but they are low risk, particularly if the code can't be unit tested.

Refresh Oldschool GUI Design [closed]

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 4 years ago.
Improve this question
I'm building Desktop Software for over 10 years now, mostly it's simple Data-Input Software. My problem is, it's always looking the same: A Treeview on the Left and a lot of Text/Data Fields to the right, depending on the type of data currently is worked on. Are there any fresh ideas how such software nowadays should look like?
For further clarification:
It's very hierarchical data, mostly for electronic devices. There are elements of data which provide static settings for the device and there are parts which describe some sort of 'Program' for the device. There are a lot (more than 30) of different input masks. Of course i use combo boxes and Up/Down Entry Fields.
Having all of your software look the same thing is a good thing. One of the best ways to make it easy for people to use your software is to make it look exactly the same as other software your users already know how to use.
There are basically two common strategies for how to handle entry of a lot of data. The first is to have lots of data entry fields on one page. The next is to have only a few data entry fields but a lot of pages in a sort of wizard-style interface. Expert users find the latter much slower to use, as do users who are entering data over and over again. However, the wizard style interface is less confusing for newer users since it offers fewer elements at once and tends to provide more detail on them.
I do suggest replacing as many text fields as possible with auto-complete-based combo-boxes. This allows users to enter data exactly the same as with text-boxes, but also allows users to save typing by hitting the down key to scroll through choices after typing part of the data in.
Providing more detail on what data is being entered would probably yield more specific answers.
I'd also answer with a question, which is to ask what your motivation for considering a change is? Like the other posters, I'd agree that there is some value in consistency, but there's also a strong value in not ignoring niggles-in-the-back-of-the-mind feelings you have. Maybe you have a sense that your users aren't as productive as you'd like them to be, or you've heard feedback to that effect from your customers, or you're just looking to add some innovation for your own interest. Scratching itches is a good trait in a developer, in my view.
One thing I'd advocate would be a detailed user study. How much do you know about what your users do with the interfaces you create? Do you know the key tasks, the overall workflow? Would you know if one task regularly consumed 60% of your users' time, or if there was a task that was only performed once a month? Getting a good sense of what the users actually do (and not what they say they do) is a great place to start thinking about what changes might be worthwhile, especially if you can refactor the task to get a qualitatively different user experience.
A couple of specific alternative designs you might like to include in re-visioning the UI might be be facet browsing (works well for searching and exploring in hierarchies), or building a database of defaults / past responses so that text boxes can use predictive completion. However, I think my starting point would be the user study.
Ian
If it works...
Depending on what you've got happening with the data (that is, is it hierarchical, or fairly flat), you might want to try a tab-based metaphor, or perhaps the "Outlook-style", with a sidebar showing the sections of an application. One other notion I've played with lately is the "Object desktop" that I first saw proposed by Scott Ambler (Building Object Applications That Work). In this, you can display collections of items, or the user can "peel off" individual records for easy access.
Your information is not enough to really suggest you an interface alternative. However, may I answer your question with a question? Why do you think you have to change it? Has your customer complained? If not, it looks like your customer is happy with the way the software works right now, thus I wouldn't change it. If your customer complains about it, he'll most likely not just say "It's bad", he will say "Why can't it look like ..." and this will give you an idea how to change it.
I once had to re-design a very outdated goods management system. The old one was written for a now dead database system, still running in MS-DOS. The customer suggested I should create a prototype how this re-implementation might look like and then he'll decide if I get that job or not. I replaced the old, dead database with a modern MySQL database, I replaced the problematic shared peer access with a client server approach and I chose to rewrite the UI in Java, since different OSes were used and this had the smallest porting costs. So far the concept seemed good, the customer liked it. However, when he asked his employees what they think about it, they asked "So far it's great, but we have one question: Why doesn't it look like the old one?". Actually, it turned out that even with all the modern technologies, they wanted the interface to exactly look and being operated like the old one. So I had to re-build a 1986 usability nightmare MS-DOS UI in Java, because no other UI was accepted.
For me it is more about a clean, usable, logical design than anything else. If your program makes sense to the user, isn't clunky and works as advertised, then everything else UI related is essentially just like painting the house. I've sometimes rolled out a new version of a program with essentially the same controls that are skinned differently.
There's a reason that you've probably chosen the tree view - because it probably makes really good sense to do so. There are different containers and controls available in the various UI libraries, depending on the language, but I tend to stick with the familiar because the user probably gets how a tree control works and how a combobox works.
A user interface needs to be usable, just don't do the misstake to change to something working to something fancy-schmancy just because it looks better (been down that road)...
Make sure that added
widgets/controls really add a business value
Make sure that the added
widgets/controls do not mess up your
architecture (too much) and makes
the application harder to
manage/maintain
Try to keep platform standards on
how to do things (for example the Vista ux guidelines)
:)
//W

How to write a spec that is productive? [closed]

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
I've seen different program managers write specs in different format. Almost every one has had his/her own style of writing a spec.
On one hand are those wordy documents which given to a programmer are likely to cause him/her missing a few things. I personally dread the word documents spec...I think its because of my reading style...I am always speed reading things which I think will cause me to miss out on key points.
On the other hand, I have seen this innovative specs written in Excel by one of our clients. The way he used to write the spec was kind of create a mock application in Excel and use some VBA to mock it. He would do things like on button click where should the form go or what action should it perform (in comments).
On data form, he would display a form in cells and on each data entry cell he would comment on what valid values are, what validation should it perform etc.
I think that using this technique, it was less likely to miss out on things that needed to be done. Also, it was much easier to unit test it for the developer. The tester too had a better understanding of the system as it 'performed' before actually being written.
Visio is another tool to do screen design but I still think Excel has a better edge over it considering its VBA support and its functions.
Do you think this should become a more popular way of writing spec? I know it involves a bit of extra work on part of project manager(or whoever is writing the spec) but the payoff is huge...I myself could see a lot of productivity gain from using it. And if there are any better formats of specs that would actually help programmer.
Joel on Software is particularly good at these and has some good articles about the subject...
A specific case: the write-up and the spec.
Two approaches have worked well for me.
One is the "working prototype" which you sort of described in your question. In my experience, the company contracted a user interface expert to create fully functional HTML mocks. The data on the page was static, but it allowed for developers and management to see and play with a "functional" version of the site. All that was left to do was replace the static data on the pages with dynamic content - this prototype was our spec for the initial version of our product. The designer even included detailed explanation of some subtle behavior in popup dialogs that would appear when hovering over mock links. It worked well for our team.
On a subsequent project, we didn't have the luxury of the UI expert, but we used similar approach. We used a wiki to mock a version of the site. We created links between the functional aspects of the system and documented each piece of functionality in detail. Each piece of functionality could, in turn, link to detailed design and architecture decisions. We also used to wiki to hold our to list feature list for each release (which became our release notes). These documents linked back to the detailed feature page. The wiki became a living document - describing our releases and evolution of our system in great detail. It was an invaluable resource.
I prefer the wiki to the working prototype because it's more easily extensible - growing and becoming more valuable as your system evolves.
I think you may have a look about Test-Driven Requirements, which is a technique to make executable specifications.
There are some great tools like FIT, Fitnesse, GreenPepper or Concordion for that purpose.
One of the Microsoft Press books has excellent examples of various documents, including an SRS (which I think is what you are talking about). It might be one of the requirements books by Weigert (I think that's his name, I'm blanking on it right now). I've seen US government organizations use that as a template, and from my three work experiences with the government, they like to make their own whereever they can, so if they are reusing it, it must be good.
Also - a spec should contain NO CODE, in my opinion. It should focus on what the system must do, should do, and can not do using text and diagrams.

Resources