Affinity Photo Workbook and Affinity Publisher

[MikeW]

Alpha testing can be valuable in many situations. It is less common for closed, proprietary software systems but it isn't unheard of. It is fairly common during testing of OpenSource software--where alpha and beta testing can be blurred together.

 

I alpha tested frequently in the later 1980s and 1990s. A couple times in the 2000s. I did alpha through beta testing for proprietary software as late as last year that ended its cycle early this year. Heck, through the 1990s into the mid-2000s I was writing an important system that went to closed alpha testing a few times. It's still up and running today.

 

One could say that while becoming more common, even Serif's open beta is unusual. They too offer up the caveat not to use it for critical work.

[R C-R]

1 hour ago, MikeW said:

Alpha testing can be valuable in many situations.

Of course. But my question is what situation(s) might there be where alpha testing of APub by end users actually would be valuable? If there are, what would be the value to end users or to Serif?

 

The Affinity apps are not open source software. They rely heavily on closed source, proprietary code. For what should be obvious reasons the developers are not going to open that up to public scrutiny. So I do not understand what the users asking for access to an alpha version expect to gain by that.

[MikeW]

I think your last paragraph doesn't have anything to do with what I wrote, at least.

 

Quote

Of course. But my question is what situation(s) might there be where alpha testing of APub by end users actually would be valuable? If there are, what would be the value to end users or to Serif?

 

Any testing along the development chain is only valuable to the company when there is quality feedback that can be verified by the company. That doesn't matter what phase of testing is involved.

 

It's just a question where along that development chain testing could/should occur to provide valuable feedback. To me, the earlier Serif begins testing outside the company as features are worked on and/or baked in, the better the product will be by the time it hits an open testing situation. This especially applies to work-flow issues--the people working on it can be too close to see what's in front of them despite the best intentions.

 

Obviously Serif disagrees with that POV. That's fine, it's their kitchen.

[R C-R]

3 hours ago, MikeW said:

It's just a question where along that development chain testing could/should occur to provide valuable feedback.

Nope. It is also a question of what an out-of-company chain of testing reveals to competitors about products in development, the expectations created about features & functions that might not make it into beta or final release versions for one reason or another, the company resources required to manage & evaluate feedback, resolving/observing any licensing conditions for third party code that might be considered for incorporation into the app, not making things easy for patent trolls, & so on.

[v_kyr]

8 hours ago, R C-R said:

... But my question is what situation(s) might there be where alpha testing of APub by end users actually would be valuable? If there are, what would be the value to end users or to Serif?...

If you make yourself a little bit smart about the general purpose of test cycles in software development, you can answer these questions yourself!

So I think the following could be of interest for the general common understanding of the different test cycles during the software development process.

• Test trategy

• Alpha Testing vs Beta Testing

• Difference between Alpha Testing And Beta Testing

• General thems in software testing

• ... and so on ...

Other related and general important thems here are also bugtracking and how to deal with it, documentation and FAQ handling etc. - All in all there are a lot of (more) things which do play an important role during the evolutionary software development process and then further the whole lifecycle of a software product.

[R C-R]

4 minutes ago, v_kyr said:

If you make yourself a little bit smart about the general purpose of test cycles in software development, you can answer these questions yourself!

Thanks, but I am already familiar with the purpose testing cycles -- how smart I am about that is a different subject I prefer not to dwell on. 

 

Anyway, as your first link (with its interesting typo!) makes clear, outside alpha testers must be willing & able to provide detailed technical feedback during what could be numerous test cycles before a product reaches beta stage; & to do so with incomplete documentation, missing features, & plenty of crashes. Finding qualified outside testers is not easy & retaining them for the duration is harder still. For products that include proprietary technology, NDA's & other legal protections are the norm & further limit the pool of suitable testers.

 

That takes me back to my question about if or when alpha testing of APub by end users might actually be valuable, & to whom. I doubt anybody outside Serif is qualified to answer that. They do not seem very keen on the idea. I assume they have good reasons for that.

[v_kyr]

It probably also depends on the experiences they might have done before here, if they are keen on these ideas or not. Or they might have not the time and manpower to setup such phases here.

11 minutes ago, R C-R said:

...Finding qualified outside testers is not easy & retaining them for the duration is harder still. For products that include proprietary technology, NDA's & other legal protections are the norm & further limit the pool of suitable testers...

Usually for such things field tests are better suited and these are mostly done in terms of selected person/tester partner programs. And yes such things do always generally run under NDA's for well known reasons, no matter if it is proprietary technology here or not. If you ever participated in such tests you know that it's more like a closed circle or group of technology partners, so to say strong usage supporters or affiliates etc. Meaning such tests usually aren't done in a public open manner! - For the later more public oriented stages, beta- and RC-tests are better suited and that's what Serif already does with their Affinity products so far!

[furtonb]

15 hours ago, R C-R said:

Alpha versions of software apps typically are incomplete & unstable. They often have only placeholder code for as yet unimplemented features that does nothing or code that causes app crashes, serious memory leaks, or file corruption when called.

 

In other words, that are not usable even for casual, non-criticals work. So why would you want to try running one if you were not a developer working on the code?

 

That's not necessarily the case. Render Legion demonstrated this, when they offered alpha versions of Corona Renderer first for 3ds Max, then Cinema 4D. They stated all the things above, and that was true: crashes and bugs were all around, but look at the Gallery, what people created with even the alpha - I'm pretty sure it helped the development to be faster and made it possible to adapt to user needs. It helped to form a very friendly and helpful community as well, which is awesome, and ties the users even closer: knowing they can rely not only on the official support, but the community knowledge too.

 

I get your side, that you don't want to release half-baked products, which crashes every 5 minutes, or makes files corrupt, but the constant postponing of release (,,it will be out when ready") makes people disappointed and just a look at how you imagine the UI, basic workflow, etc. (which features are "kind of" working), would ease us. Knowing that when ready, it will be an easy switch already. Early adopters (alpha-beta testers) will be able to jump on day 1, without hesitation, because they are familiar - with the limits as well, of course. I think the latter aspect is more important: get to know the software and help decide, which features are awesome, which needs further thought or which ones are the most requested ones - not speaking about the wide range of different uses on different hardware, which can lead to a ton of useful crash data.

 

So yes, I'm not a developer, just some random guy who had a very good experience with another perspective of software development.

[InfoCentral]

Received my Photo Workbook yesterday from Serif!

[rgcn]

I love the Photo Workbook great work everyone involved!

 

Rob