Software Quality In a Non-Software Company?

Nicros writes "I work for a publicly traded biotech company that happens to write software that is, in fact, kind of critical for the business — without it no data would ever be read from our instruments, and no analyses would be performed on that data. The problem is that as a 'biotech' company, we are not taking software quality seriously. We have no senior management with any history of commercial software development — our C level has really no clue whatsoever what software really is, much less what is going on in software development. All of our quality processes are related to manufacturing our system (not software), so we are constantly forced into ad-hoc development since there is no real process for our development. Repeated requests to hire someone with some real commercial software development experience have gone unanswered. I have been to the CEO directly one-on-one and although he agreed this was an issue, thanked me, and said he would look into it, that was the end of it. He has bigger things to worry about. So the question: Is this just a fact of life and I need to deal the best I can? What else can I do to get some attention on software quality in the company?"

Comments

Practice What You Preach

[ilovegeorgebush]

You're obviously fighting an up-hill struggle. Going straight to the CEO is both a good and bad idea - if it works you'll get immediate affect, but it's likely to be ignored.

You need to argue this case as much as possible. If you're the developer, or in charge of development, enforce decent developmental practices and ensure your estimates include them. Err on the side of caution. Take an estimate and double it. Managers talk money, not standards, so you'll have to hit them where it hurts.

Otherwise, is there anything off-the-shelf that could alleviate some development?

Re:Practice What You Preach

[b4upoo]

Between being right and being popular with bosses being popular wins out every time. Pestering them about better ways to do things is not a great idea. Play golf with them and never mention a darned thing relating to work and if something does crop up then make them think it was all their idea all along. Money shall flow to you as well as job stability. Make sense and be logical and you might as well start a job search.
          Thia may not apply in other nations but in America it works every time.

Re:Practice What You Preach

[Hotawa Hawk-eye]

Your statement "We have a problem, this is how other people solve it, and this is what I will need to solve it. Give me the budget and I'll solve the problem." is close to complete, but you're missing one piece.

"We have a problem, this is how other people solve it, and this is what I will need to solve it. This is what it costs us _not_ to solve it. Give me the budget and I'll solve the problem."

If you can show that the software development 'process' currently in place is costing the company $N a month and you will need to spend $X to improve the process, if you're going to be developing software for more than (X/N) months, it'll be more cost-effective to fix the process.

Re:Practice What You Preach

[electroniceric]

I also work for a biotech (... this my other brother Darrell), and we're facing the prospect of FDA regulation as a device. So we're presently working our way through ramping up a formal (21CFR820) Quality System now. My boss happens to have been through this before and to be a pretty effective evangelist for the FDA's Quality System methodology, which is required for all medical devices and drugs. As he says, a Quality System is just putting to paper all the things you should be doing anyway. So one place you may want to start is by discussing the utility of complying with FDA regs with regard to software.

My boss also notes (on any occasion where there's an opening) that when the FDA introduced design controls, most companies complained they were going bankrupt (as companies are wont to do when regulation, merited or unmerited, is proposed). But when the FDA went around doing their roadshow to show that they weren't just making rules without listening to industry, people from the device companies gradually started to get up and explain how using a Quality System actually lowered their costs and decreased their time to market for revisions and product upgrades.

So as an evangelization tactic I'd look on the FDA's site for guidances relating to the introduction of the Quality System Regulation. For example, this guidance on general principles of software validation is pretty good. If you mentally translate into software industry language you can really see that they're trying to get you to do better engineering by thinking and documenting early, really getting straight what the software is trying to do, and being structured about showing that it does what it needs to do. The truth is that despite the startup effort of introducing documentation and procedures, controlled engineering methodologies work way better - they reduce requirements failure, increase code quality (and more importantly, design quality), and - though developers start out hating paperwork - even make the developers happier because more code works and sees real use. If the company plans to be in the software business for the foreseeable future, it's almost certain that the effort invested in good software practices will pay huge dividends down the road. The key is point out that quality is not an esoteric consideration, it's a driving cost and business risk consideration. Sooner or later the cost of low quality software shows itself.

One thing I will note is that the QSR is pretty waterfall oriented, both because it predates the formalization of iterative/agile methodologies, and because it's written for engineering of physical boxes that have to be released to manufacturing (which implies a fair bit of waterfallism). Part of our effort is to practice iterative development methodology while documenting to the FDA's standards.

Anyway, take a look at all that.

Anarchy is an opportunity

[Hal_Porter]

It sounds like in your company there is no one doing this job. You've talked to the CEO. Get him to make you VP of software and tell him you'll solve the problem if he gives you responsibility.

Anarchy is an opportunity for the ambitious and unprincipled. Take it and make yourself software Czar.

Re:Anarchy is an opportunity

[pieterh]

Agreed. Don't raise issues for other people to solve, you are just labeling yourself a trouble maker. Raise issues, attach costs to them, and then present yourself as the person with solutions, and ask for budget to solve them. Make a proposal with figures, planning, clear savings, and get approval. Then hire and build a competent team and/or find a good subcontractor. Use open source where possible to save costs. Report your progress and ensure you get budget every year.

Think of ways to turn a profit from the software. Maybe it can be licensed to other firms? If you can earn revenue you will suddenly become much more valuable.

Problem is: you will stop coding and become a manager. But if you do a good job, you can get power in the firm.

If you present a good plan that will solve real problems for the company, and you are not given the green light, then look for another job. If/when things go bad, they won't thank you for it.

Re:Anarchy is an opportunity

[southpolesammy]

Careful what you wish for though....

The flip side of becoming a point of authority in an environment as this is that if/when code defects bubble to the surface on your watch that result in a major hit to the company's bottom line, you may need to have a thick layer of asbestos underwear on in order to prevent the blame game from claiming you as a victim.

Right now, you've identified the problem for your mgmt and have suggested solutions, but you're not yet responsible for the implementation of those solutions. Becoming the VP of such an org not only makes you responsible for the fixes, but also directly accountable, possibly including from a legal standpoint. In other words, you'd better hope that the bugs in your software don't have the potential to cause medical or financial harm to your customers.

Re:Anarchy is an opportunity

[houghi]

Also do not forget to get some customer input. It will be extremely hard to change anything if your customers are perfectly happy with what you have. For you it might be bad, but perhaps for the customer it is 'good enough'.

Doing the manager thing and figuring out the numbers might even lead you to the fact that changing things would cost more, while not gaining extra income or saving anything.

So it could be that it is just a 'nice to have' and not a 'need'. You will need to prove it makes money, otherwise they will not do it.

'Making money' can also mean that your customers will be happier, or they will call you less, saving money, because you can do other things. It can also happier customers, which might be important if you measure customer satisfaction.

Basically you need to sell the idea and the company needs to buy the idea. If you can agree on a price where both parties gain in the deal, then it is good. If one of the parties does not gain anything, then it is a no-go.

This is true for every change in whatever it is you want to change. From the color of the toilet paper to the closing of a factory.

Re:Anarchy is an opportunity

[kitgerrits]

If you want to keep yourself safe, document and report all 'blind spots' in your method.
Make sure you present an overview of what you can control and what you cannot control.

If management does not agree with a certain blind spot, show them the resources required to cover that blind spot.

You cannot have bug-free code without strict rules and a literally astronomical budget. (and even NASA has had a few bugs)
What you can do is prevent embarassing/dangerous bug from making it into 'production software'.

Opinion from the outside?

[DerCed]

Could you propose to hire a software test consultant for a day or two and let him point out serious quality issues (data integrity, security, correctness..)?
A serious, alarming report by an external software test professional could help reinforcing your requests?

I have the same problem

[_Shad0w_]

I have the same problem where I work, the problem is I am the dev with real commercial experience; I just can't convince them that we need to do things in a manner that I would consider correct - it's all ad-hoc development and it's all driving me nuts.

The problem is, if our software doesn't work correctly, then the data we collect and process using it becomes screwed up, which is a major issue for the core business - data is our crown jewels.

My current solution to the problem is looking for a new job in a company that actually takes software development seriously. I just can't see any way of getting things here working the way I want. There wasn't even any revision control in place on the source code when I started.

The problem I'm finding is that the lack of structured development and design here is actually beginning to hurt me professionally: prospective employers, who have software development as a core aspect of their business, actually ask about this kind of thing. If you're looking to hire someone who takes their profession serious, for god's sake make sure they're actually going to be able to do their job - otherwise your company is just going to turn in to a blot on their CV.

Mixing of two mindsets

[slipnfall]

Hi OP,

I'm a developer/Engineer for two biotech companies: one a small startup, with me being the only part-time employee. The other is a large DOD-backed institution. I can tell you that in the short time that I've been there, it has been a frustrating uphill battle to instill an Engineering/Developer mindset. While I firmly believe Scientists and Engineers seems to have a similar approach to work, it's interesting to see how passive the science-minded folks are towards hardware/software advancements. They are only concerned about how many protein cells it can accurately count, or whatever. There is no interest in what goes on 'behind the scenes', and consequently, what goes on to get there.

There are absolutely no Engineering controls in place at either Employer, and software development is as you said: made for the moment. Personally since I am the one-and-only, I find that I just have to do the best with what I have. I comment and doccument well, keep a code revision repository, and do my best(within reason) to make sure someone else can pick up where I left off.

It won't be my problem if/when the day comes I leave, but at least I'll be able to sleep at night.

Some hints for your situation

[Apogee]

I'm in a quite similar situtation, and perhaps I can provide a few hints from what we're currently doing.

I'm working for a relatively well-known institute in academia (biotech field), with a group that among other research projects, also provides web-based services to the research community. Funding is partially tied to the operation of the services, so there is actually enough pressure to make sure that they work and work correctly at all times.

Still, until about a year ago, development was very ad-hoc, in a mix of languages, and with many "islands of knowledge", where some parts of the system were only known to one post-doc, and other parts could only be fixed by the group head (who, as they are, was usually busy with many other things...). After some hard times and near-misses, we started looking around for ways to improve our development.

I was quite attracted by the ideas of Agile, and I believe that they're a good fit to the kind of processes you find in science, as well as in software engineering. We initially had a professional Scrum coach come in and talk with us about software development practices, and then decided to apply Scrum to our processes.

It's now a bit more than 1 year since then, we're still using Scrum with a few adaptations to fit the academic environment (we're also using Scrum for projects that are really science and research, not software development). In a recent secret poll among the team, Scrum got high marks for making the team more productive, and for creating an environment where code and knowledge is shared. People are happy with the structure that Scrum provides, and we always know where the project stands. Incidentally, we also write better software faster.

But we're still improving the way we work. The transition is slow and painful, and we're only slowly adopting things such as test-driven development, automated builds and pair programming. In my experience, there's a lot of resistance against these "newfangled" methods in the academic culture, especially that of people who weren't trained as software engineers, but rather as physicists, chemists, biologists, but now find themselves producing software.

Some hints on what I've found useful in re-shaping our work environment:

- You can't change the whole structure in one day. Get permission to run a small, isolated project in "the new way", and use this to demonstrate the advantages. Remember, there are many metrics for success: Code quality, timely delivery, not having single points (persons) of failure, as well as team velocity and personal satisfaction. Try to make a case from this small project (and gain experience while doing so), and then grow it out slowly.

- I would not advise to do some clever "breaking the build, and thus showing everybody how fragile the system is" exercise. This may not be seen as constructive.

- Instead, provide convincing evidence by example that your way is more productive and more certain. Bugs that are fixed stay fixed, and don't creep in later again. Timelines are better kept. That sort of thing...

- If you can get someone in to talk about the current best thinking in software development, do so (someone else mentioned this already). It's good to hear an outside opinion, and to understand that these practices are not theoretical but used by large companies world-wide.

- I found Joel Spolsky's 12-point assessment very useful to find out where your organization stands: http://www.joelonsoftware.com/articles/fog0000000043.html ... These are also good points to whisper into management's ears.

What exactly means "Non-Software Company"?

[Tanuki64]

I have worked in enough software companies to know that they are not necessarily better.

Open source

[Confused]

• Your software doesn't make money for the company, it's just producing costs.
• Your customer need the software to use your stuff.
• From your description, your customers might include quite a few very clever ones that constantly try to push the limit of your systems and thus damning your software to eternal hell for its shortcomings.
• Any help you can get to develop it would be welcome, although don't expect your development costs to go down.

This sounds like the perfect scenario for open sourcing your software with you as the main developers maintaining it.

For the regular users, nothing much will change.

For the power users, those most likely to complain, this will be a tremendous benefit. If they don't like it, they have the possibility to improve it. This often reduce the number of problem reports and increase the good problem reports from your knowledgeable customers. Sometime you might even get useful patches, that save you some work. If you're really lucky, you might get a few users who start to code enhancements.

It also might generate some good-will towards your company and ease the integration of your bricks with other solutions.

What has this all to do with software quality? With your software out in the open, quality problems tend to be treated more like bugs that will be fixed as fast as necessary and possible and you get a better feedback where work is important. Making the software and drivers open source won't save your company any money, it won't cost more either, but it will improve what you get for your effort.

Missed the obvious solution

[johannesg]

Make it fail. Make it fail spectacularly, to the tune of millions of dollars. That will certainly get the CEO's attention, and he will be sure to take measures that will stop such failure in the future. Of course, I can pretty much guarantee that you will not like his solution, but software development will be much more professional afterward.

If this is not what you want, ask yourself what you actually want to change. You do know what you want to change, right? Discuss those things with colleagues and managers, then formally propose doing them.

I'm guessing you probably want a more structured development process, with better-organized change requests, and at least some semblance of formal testing. That is very, very hard to set up, because it also requires the help of your users, and they don't care about software, they just want to have their problem solved. If this is the case, always remember that you are there to solve their problems, but they are not there to solve your problems. In other words, don't force them into a process they don't like. You might do better if you can show an advantage other than "it makes my life a little easier".

If all you want is a bugtracker and a version control tool, just request a budget of about $2000, then buy a Dell PC with Linux and install Bugzilla and SVN on it. That will set you back $400 or so, the rest of the $2000 is to show that you are a business thinker and did not forget to include installation time ;-)

If you want to institute Methodologies (like extreme programming or similar), good luck with that. It will probably end in your colleagues defenestrating you...

Re:What the Hell are you talking about?

[SpinyNorman]

Go read up on the Capability Maturity Model (CMM) or ISO 9000 and come back when you have a clue.

You don't even need to formalize the process to that extent to make leaps and bound improvements on the hack-it-together and test it approach you are suggesting... At a minimal a decent software development process should have:

Requirements specifications & reviews
Design specifications & reviews
Test specificiations & reviews
Codng standards
Code reviews
Source control
Regression tests
Functional tests
Load tests

Re:Here's a revolutionary idea

[indifferent children]

How much combined experience does the management team and board of this company have?

This argument is also known as "The Enron Gambit": those wildly successful guys who are raking it in hand-over-fist must know better than those of us who think that their business model makes no sense. They sure showed us.

Use the right language

[ebbe11]

...and that language is money.

Find out how much it will cost the business if the software stops working. Estimate the risk (number between 0 an 1) of this happening. Multiply these two numbers. The result in dollars is the amount of money your company will lose with certainty. Not maybe, with certainty.

Garbage in, Gospel Out?

[Anonymous Coward]

Unfortunately, software quality isn't even on most companies' radar. Until it exposes them to major losses like a balsa-wood skyscraper built next to the airport on the shores of the Petrol Sea.

Software has the disadvantage of being intangible and we all know that "Any kid can write software".

Any kid can bandage a cut, but that doesn't mean you want that kid doing a colonoscopy on you.

At some point the software industry is going to need to establish itself as a rigorous practice with rigorous standards. Not some silly cert that says you know Language-of-the-Week, but something along the lines of GAAP for accountants. I'm not holding my breath, though.

IF you happen to have - or be able to cultivate - the right social skills, take an active role. However, despite what the "don't like it, get-entreprenurial" crowd asserts, there are those of us who'll never be able to tolerate forcing their introverted personalities to assume an extroverted task on a long-term basis even with the best of counseling, self-help and medications. It can be wearying and it steals time energy from what we can do that the extroverts can't.

So if you aren't socially adept and don't see yourself swimming through office politics like Nemo, the best advice I can give is keep your resume up to date and network to whatever degree your social skills allow so you can bail before the tower collapses.

Then again, you can be Monty Hall and still come out of the losing side in the office, so keep the resume up to date anyway.

Re:As a C-Level for a Software company

[roman_mir]

Do not bring up agile in this situation. You want to push them into a paradigm that is structured towards responsibility, not something that allows development to wash their hands from anything and just blame business. In the situation described in this story the best way to go is by setting up a structure that forces a couple of things: documentation of requirements up front, system and design specs, phased iterative development, unit testing of course, QA department, responsible management. I know it sounds difficult, but you have to work towards it, nothing is easy.

Get a test audit

[ArtistFrmrlyKnwnAsAC]

I write software for clinical researchers. From the summary description, sounds like the company has software that is part of their clinical quality system which is not being tested/validated. If the description is correct and the software is actually part of the clinical therapy they're selling, they should have an external auditing agency take a look at it before the FDA does.

The first time one of their products is audited by the FDA, the warning letter they receive will communicate to management exactly what is lacking in their compliance with FDA cGMP. Unfortunately, everyone from regulatory compliance down to the lowliest coder who had something to do with the products in question will share in the group spanking (been there).

I'd be shocked if they didn't already have a relationship with an auditing company--unless they're the tiniest of startups. If they do, the submitter should look through their last audit summary and see if anything has gone unaddressed, and if the scope of the audit matches the submitter's idea of the actual quality system for the product(s) in question.

Re:Plant a bug

[ilovegeorgebush]

It's hardly going to bring down the company (unless the company was in such dire straits that anything small could bring it down, in which case it's about time to leave), and may bring to the attention of the upper mucks of an important issue.

There are better ways of doing that, such as persevering, practicing what you preach (good development standards and approach), being firm with management about the issue.