A recurring theme of mine is building flimsy analogies between software processes and other processesTo give a very recent trivial, but valuable to me, example: I use filters aggressively for sorting much of my mail into a hierarchy of folders for batch reading or archiving. However, there's a category of mail that I want to sort manually, not least because it forces me to read it. Over time, sub-categorisation within it has become more complex but for now I still see value in preserving it. This means that my manual filtering progressively involves more and more dragging and dropping through hierarchies of folders. This is inefficient and irritating.
In this example it's no great conceptual leap to think of some kind of analogy between the mail client folder structure and a file system. When I move an email from one folder to another I am performing an operation similar to a file being copied in a file system. So, in a file system, is there a way to reference deep folder structure conveniently? Well, yes, there is: symbolic links, also known as shortcuts, can preserve the structure (which I value) while making "virtual" folders available anywhere (which I want for efficiency).
A quick search for "Thunderbird shortcuts folders" throws up Tabbed Folders, which gives me what I need and an immediate productivity gain. By asking how the analogy domain would resolve the issue, we can expose a solution in the problem domain.
This particular example fits well with the Comparable Products consistency heuristic from HICCUPPS:
We expect the system to be consistent with systems that are in some way comparable. This includes other products in the same product line; competitive products, services, or systems; or products that are not in the same category but which process the same data; or alternative processes or algorithms.but an analogy can be to anything, software or not, and be useful. I invoke analogy frequently - I might simply ask "what is this like?" or "what are synonyms of the important elements?" - not just when actively testing but also when thinking through all sorts of problems. I find it can help to move a thought on, focus it, or broaden it out, or reject it. It may be a piece of temporary scaffolding to a longer train of thought, or a seed for lateral reasoning.
Even so, despite the value that the example analogy gives, many aspects of the two systems are not analogous at all. For instance, I am unlikely to want to my mailer to have the kinds of complex read/write/execute permission permutations that most file systems do. Like other kinds of heuristics, analogies are usually incomplete and it's important to bear that in mind.
1. That's just a joke; it's actually completely incoherent. No but, really, we do get on well although the last time I saw his wife, her opening conversational gambit was "How's your shit blog going?"
When I have questions that require a response from others, I often try to provide a selection of answers too. For me, this helps to think through what I might do next - perhaps I can discover the actual answer by experimenting to see which of the answers is likely, or maybe an answer prompts another question, a more interesting question, one which requires investigation itself.
If I still go ahead and pose the question, having some potential answers alongside it shows that I've thought about the issue and gives the answerer some context. Also, in my experience, people are often more able to refute an assertion than generate an answer from cold, so my answers spur their thoughts too. (And by the way, don't forget to answer others' questions.)
When I responded to Phil Kirkham's STC forum post asking for three pieces of advice for a newbie tester, I included answering your own questions. In fact, the advice I gave for newbies forms three quarters of the advice I give to myself during a mission:
Ask questions - including of yourself
Answer questions - including your own
Act - on what you now know, or to find what you need to know
Again? - there might be more to do here
In a section on the design of a clinical trial (where interventions - such as drugs - are being compared for efficacy in treating some illness in humans), the term equipoise was introduced. The definition given is:
Substantial scientific uncertainty about which treatments will benefit subjects most, or a lack of consensus in the field that one intervention is superior to another.and the course notes say:
A state of equipoise is required for conducting research that may pose risks to research participants.For a clinical trial to be in equipoise, investigators must not know that one arm of a clinical trial provides greater efficacy over another, or there must be genuine uncertainty among professionals about whether one treatment is superior than another. Equipoise is essential for obtaining generalizable knowledge. If a clear and agreed-upon answer exists, asking research participants to assume the risks of research that will provide the same information is not acceptable; no new knowledge will be gained from the study.According to Wikipedia regulatory bodies do not agree on the utility of the concept, but the pro argument runs like this: the need for equipoise arises from a potential ethical dilemma. For example, is it ethical to run a trial in which the researchers are certain that one treatment is substantially better than another, i.e. where some of the participants with an illness will almost certainly receive other than the optimal treatment? Or, even if an experiment started with genuine uncertainty, imagine that the researcher has strong evidence that one drug is dramatically outperforming the others under investigation before the experiment end. Should they switch all of the participants to the better treatment? If so, at what point?
Opponents of the use of equipose argue that it mixes up concepts of therapy and research to create the ethical dilemma. Only by regarding the investigators as providing treatment to the subjects, is there any medical-ethical obligation on them to provide the best treatment. In research, the greater good of the population as a whole might outweigh the needs of a specific individual - presumably up to some point where the treatment under test is proving actively detrimental to the subject.
And so to software testing. We're in the risk business too. Could equipoise be of use to us?
At one level there's the idea that peer decision-making is appropriate under some circumstances. I think we'd probably all accept that. Which of us doesn't sometimes ask a colleague or the community for advice or a second opinion? As stated here, though, the need for equipoise is predicated on the possibility of risk to participants. Glossing this as simply risk to someone who matters, there's unlikely to be a test that doesn't pose some risk if only because by performing one test, in a fixed budget, there's almost certainly some other test you didn't perform.
Continuing that thought, the notion of equipoise itself doesn't seem to take account of the level of risk or the cost of an experiment. If it's cheap to perform a test with low risk to the people who matter, perhaps the level of uncertainty required to motivate the test should not be the same as an expensive, high-risk test.
The text I quoted includes the phrase "if a clear and agreed-upon answer exists ... no new knowledge will be gained from the study". When trying to decide whether to perform a particular test, we'll ask ourselves whether we think we might find something new - a debate that frequently squirts out of the side of discussion on regression testing such as this one yesterday on Twitter - or, perhaps better, whether the cost (including opportunity cost) of running the test justifies the risk that we won't generate novel results.
In the general case, trials are long-running with reasonably specific aims couched in terms of hypotheses. The investigators will designate confidence thresholds above which an hypothesis will be said to be true. For example: drug A is better on the general population than drug B because the statistics we have done, based on the attributes of the patients that we measured and our sample set, which we selected carefully to be representative of the population and randomised for the trial, which was also double-blinded, were significant at the 95% confidence level. Most software testing isn't framed this way or, at least not formally. Much software testing is framed in the form of a question and a binary opposition - does it X (by whatever criteria)? Pass or fail.
Looking for insight I tried analogy (leaving aside any scenarios where the software is medical in nature).
The researcher is probably uncontroversially the tester. The subject of the experiment has a couple of obvious candidates: the software and (by proxy) its stakeholders. And if the subject is the software then the risks might be to do with performance, robustness, scalability, functionality and so on. If the subject is the stakeholders, then risks are to the value that the stakeholders obtain from the software.
What is analogous to the trial? Is it simply a test? Or is it a sequence of tests? What if the trial is asking multiple questions? Is each of them a test? In the clinical trial, would a single dose of some compound be a test? What about if data was collected and analysed after that dose? And how about the intervention? Perhaps that's somewhat like test data, or the steps used to perform the test, or a configuration for the system under test?
Under what circumstances could we have an ethical dilemma? In the trial, the health of the subjects and the obligation or otherwise of the researchers motivate it. In our scenario, what would the health of our subjects be? Well, a program might perform better (be faster, use less resource or whatever) in some contexts. A stakeholder might be happier if the product can do some things rather than others.
Which brings us back to the intervention - under what circumstances could a test provide these "health" benefits? Well, configuration options could tune a product, or a particular environment could allow it be more or less performant. In the case of stakeholders, perhaps their confidence is boosted by test results. Or maybe we need the intervention to be something that could change the product and more directly affect value, perhaps a test and corresponding software change, if required?
So are we in a position to recast equipoise and maybe find a potential ethical dilemma? Here's one attempt:
Equipoise is substantial uncertainty about which tests and corresponding fixes will benefit the software most, or a lack of consensus that some test/possible fix cycle is superior to another.A state of equipoise is required for conducting tests and applying any corresponding fixes that may pose risks to the product's performance.Casting the product health in terms of performance, there's a possible facsimile ethical conflict. Imagine an experiment on a complex system with many configuration options. If some combination of settings is found to tune the system for incredible speed and low resource usage then we might be tempted to immediately apply that configuration elsewhere. However, if the result was found in the test lab and then applied to production instances, it's not a dilemma because those machines were not subjects of the test.
To have the dilemma, we'd need to be applying it to the test machines. So perhaps we're testing for ways to improve the reliability of the test machines; we find that one particular GPU provides sufficient uptime. Would it be a dilemma to simply stop the test and fit that GPU to them all? Perhaps we're testing in production when we find the magically performant setting combination. Perhaps that's closer to the human case - should we let some of our customers continue on the non-performant settings until the end of the experiment?
I don't think I've convinced myself that there's much of value to testers in the notion of equipoise. But on balance I probably need a peer consensus. What do you think?
Me: Done, run this for me!
Perl: No, what about X?
Me: Ah, yeah, perhaps this time.
Perl: No, did you look at Y?
Me: And now?
Perl: No, have you seen Z?
Ring any bells? I try not to be the interpreter these days, by avoiding the one-dimensional rejection that invites the instant edit/retry.
Fortunately, in the blogosphere you can't hear your readers scream, so here's an FCB of heuristics and aides-memoire that I use when setting up processes, guidelines, checklists and the like. They're most appropriate when intended to be used on projects with multiple people, probably across teams, to manage collaboration on something that is considered valuable by a person or people who matter.
There are usually a handful of roles:
- customers - the people who want the process, perhaps with quite specific demands
- owner - the person/s responsible for setting the process up
- manager - the person/s responsible for ensuring it is used as intended
- agents - the people participating in the process
- The customers are not the only stakeholders. Anyone working with the process and anyone who interacts with it has a stake in it.
- Be clear to yourself and the stakeholders what problems you're trying to solve or avoid and what goals you have for the new process.
- Try to make the first implementation something that you think is viable.
- Consider basing it on an existing process if there is one. (Either an iteration or a rejection of it).
- Especially at the start, try to involve sympathetic staff, champions of the project, or those with vested interests in seeing it succeed
- Try to make the process as close to the minimum you're prepared to enforce as possible.
- Try to avoid the temptation to add in reports, requirements, checks and balances that you and/or your customers don't really need.
- Anything that you know you'd bypass in an emergency is a candidate for cutting.
- You will probably have to enforce every step you define.
- Be prepared to do that.
- For your process to work for your stakeholders, you need your stakeholders to believe that you will make it work for them.
- Do eat the dog food. Be an agent in your own process for some of the time.
- Don't drink the Kool-Aid. Be prepared to question the process if you feel your customers have it wrong.
- If you design the process it's more credible if you're also the owner and manager (at least initially).
- If you're the owner it's more credible if you're also the manager (at least initially).
- There will be questions. Deal with them in a timely fashion.
- There will be objections. Deal with them in a timely fashion.
- Changes may need to be made. Deal with them in a timely fashion, and ensure all participants are aware of them.
- If documentation is needed keep it up to date.
- Make sure that the people working in the process can see that you are keeping it up to date.
- Solicit criticism and comments and take them seriously.
- Make it clear that you will make changes if anything is not working (well enough).
- Actually do make changes if it's not working (well enough).
- Be clear about why you're not making changes if you decide not to.
- Don't necessarily stop thinking about improvement when the process is bedded in. For instance, once everyone understands the process are there bits you can safely remove or automate?
- Initially, monitor very closely.
- If you can test it out on a small scale before putting it into production, do so and gather feedback on it from all participants.
- Pay particular attention to interfaces. For example, where does control pass between two parties? What material is required at that point? What format? What other conventions?
- Make templates for stuff that can be usefully templated. Consider this particularly for critical stuff that, if missing, would block a downstream stage.
- Don't demand templates where there's no need or where it would stifle creativity or productivity.
- Look for standard approaches/tools where standards can be useful. For instance, what can you do to make sure that project time is not spent learning how to work in a project?
- Where possible, use tools to report status, provide a framework for moving a process through whatever stages you have.
- Think about how you can tell the process is working? is there a metric you care about?
Too much information? Could I distill all that mess into some kind of compact advice weapon that's more direct than the full FCB? Sure, here's what I like to refer to as an acronym-based surgical strike (ASS):
- Engineer: to make it efficient, smooth, workable.
- Propose: to seed discussion, and then gather and act on the feedback.
- Impose: because sometimes, someone has to have the final say or force the right or required thing to happen.
- Care: about getting it right for all of your stakeholders.
And on writing itself: if you've never coded then writing for an editor is a reasonable approximation to the implement/bug report/fix cycle, with you as the engineer. Do you take criticism as well as you give it? Can you learn from that? What's stopping you?
Perhaps I just needed a drink, or maybe I've been at this game for too long, but my initial instinctive interpretation of this was that time was being called on alcohol. Is the space of all possible locations really covered by the restriction? Surely that can't be what it is meant to mean?
For a bit of seasonal fun, I set my team and the Cambridge Software Testing Club Meetup this challenge:
In my imaginary world (one where the compendium of value-for-money amenities at motorway services is a good deal larger) Moto have asked you to review their signage. Test this particular sign for them. You have 15 minutes to list questions, comments, interpretations, suggested experiments and so on.
I've recorded a few of the suggestions in the comments here. Feel free to add more.