Author Topic: Microsoft repackages apps with a telemetry .NET wrapper  (Read 2880 times)

0 Members and 1 Guest are viewing this topic.

Offline bd139

  • Super Contributor
  • ***
  • Posts: 23045
  • Country: gb
Re: Microsoft repackages apps with a telemetry .NET wrapper
« Reply #50 on: May 01, 2024, 05:30:58 pm »
Your reasoning is detached from the outcome.
That’s a rather lofty accusation.

I worked in the software industry for years, and at a usability agency. I have relevant, real-world experience with this, and am not the deluded simpleton you essentially accuse me of being.

I'm not making an accusation. I'm just stating that the data and the intent do not always end in the conclusion that people think that they do. That is mostly because people don't know how to do a proper analysis of anything really. This is not specific to this but the claims of a tangible outcome are vastly overstated.
You didn’t say “people”, you said me, so it’s hard to not see this as you basically calling me incompetent or stupid.

This is your extrapolation of my point. Perhaps I should define it better rather than let you talk yourself into thinking someone called you stupid or incompetent.

Your reasoning is accurate and fine. But external concerns, aka engineering teams tasked with implementing change, tend to shape the outcome away from the intent. You end up with something half way between where you are and where you intended to be most of the time. Which is sometimes worse than either.

And on top of that the analysis being done needs to be done without bias and when there is a sales and continuity bias or commission involved, this affects outcomes.

Fundamentally it would be better to hire you in house to slowly shape the product than it would be to advise and leave. Which is where we ended up. Well we didn't hire you but probably would have if you were there at the right time with the right portfolio.

I would suggest that the "usability agency" is a considerable bias as well based on my other comment. Literally there is no business if you tell the client not to touch something, so the default state is that a change must be made otherwise there is no report to make.
That’s on you. If a client wants to test the existing state, an agency will do it. If you weren’t getting this, it’s because you neglected to ask for it — or more likely didn’t want to pay for it when offered.

We paid for it and did not get it. Twice.

The reason that we stopped hiring agencies to run user studies for us is that at no point did anyone run a baseline analysis against a null hypothesis i.e. no change. I got into a hefty argument with a consultant over this who said that a change is 100% necessary, without providing any evidence and before the study was run. That is a complete lack of objectivity, intellectual and professional integrity in the industry.
Then you were dealing with a REALLY shitty agency. The one I worked at absolutely did not work like that.

Many of our projects began with testing the status quo to figure out where the problems are. (One memorable example was a health insurance company’s enrollment forms: an old lady was doing the test and got to the question “have you consumed cannabis products within the past 60 days?” to which she said out loud “no, but I could use some right about now!” 😂)

Don’t extrapolate to the entire industry; there ARE honest players. Just avoid the ones in any way affiliated with big management consulting companies! :P

I will also add that a LOT of the usability experience I got was at the software company, where it was all being done for the benefit of our own end users.

We dealt with multiple agencies. In the end we hired our own people and do the work internally which is the point. The motivation and outcomes are different. As you say where your experience came from.

Personally I think that's a shitty feature because it doesn't work like anything else in the rest of the OS or any other software. It's literally an edge case coded into the office UI runtime.
Jeez Louise you’re rigid!

TONS of software uses custom controls to do useful things. Done well — like here — they are unobtrusive and doesn’t interfere with existing workflows.

On Windows, the basic UI widgets are MUCH dumber than those on macOS, so Windows developers are more or less forced to use custom widget extensions or third-party widget libraries to get more advanced behavior.

Well it depends which layer of windows UI you pull in? win32 / MFC / ATL / WPF / probably some other layers I've forgotten...

But yeah I agree there to some degree. But the clipboard itself will always paste the lowest common denominator of representation of the contents of the clipboard that is supported by the destination.

Compare to "Paste and Match Style" on macOS which is system wide.
Yep, it’s a great feature. But it can’t work everywhere. The document object models of the Office apps are WAY more complex than even macOS’s basic widgets. macOS and windows both use RTF as their basic formatted text format. Word, for example, does not. It’s a stylesheet-based system with nested stylesheets which apply to different levels of document structure, so merging them requires intimate knowledge of said structure. An OS-level command can’t do that.

I know intimately how Word works. I wrote a very comprehensive document authoring system a few years back that was a Word add-in and server-side generator that used OOXML. As for the clipboard, check winuser.h - you can register your own clipboard formats which is how it works when you copy structured data from one instance of winword.exe to another. Actually it's a serialized chunk of the internal COM model still for compatibility reasons (say you want to cut and past into Minitab or something horrible and windowsy). The default formats are indeed plain text and RTF but you have to ask a question here:

What has that got to do with things?

The point is there is a standard on macOS and there is not on Windows. It's literally a UI standards issue. And because it wasn't specified then the Office UI team decided to go their own way as always.

Consistency and discoverability is an important part of usability, no? Leaving things in the same place is a very simple way to make them discoverable.

I don’t disagree in principle with that statement, but maybe I’m just not quite as jaded as you.

It's not really jaded, but experienced.
I’m experienced, too.

You’re jaded from negative experiences. You accuse me of having bias, but so do you. I’m sorry you had very bad experiences with agencies, but not all are like that. Some are comprised of good, honest people who do good work that truly benefits their clients.

FWIW one reason I left the computer and usability industry entirely is because I got jaded — about clients. Too many times, after properly studying the problem, we’d design and test a really good solution, but the client would then not implement it as designed (often because they didn’t want to put in the effort on their backend systems to actually support the interaction design they’d been seeking). The half-baked implementation then would introduce new problems, which the client would tolerate for a few years before undertaking another website redesign, throwing out all prior work and starting over again rather than improving the implementation. So as a usability engineer, I rarely got to see my actual designs implemented, just pale, unsatisfying ghosts of them.

QED on the first point I was trying to make. "Pale unsatisfying ghosts of what you designed".

You have to operate in the constraints of reality which is run by cheap, incompetent idiots.

It's better to be embedded in the org and move chess pieces slowly than try and use a consultancy to get the issues you needed sorted.

Put it this way, who's the last person you go to for financial advice? Actually a financial advisor. Why? Well it turns out that they have two principal objectives (a) earning commission and (b) reaching sales targets. That gets you a mediocre outcome. What gets you the best outcome is developing an understanding of the domain and the problem and that comes from a proper study and analysis, not witchcraft and hope for a fixed price.
Well what you’re describing isn’t a financial advisor, it’s a broker or salesperson, who at a big financial firm may well have the title of “financial advisor”.  But there exist independent financial advisors, and their advice, as I understand it, can be quite good precisely because they aren’t selling you any of the products they advise you to buy (or not). The flip side is that their services aren’t free.

You get what you pay for: a free financial analysis or free usability review is basically a way to get you in the door so money can be made on you some other way. It sounds to me like you may not have a good feel for this.

As you understand it is completely wrong. I wouldn't have made the comment if I didn't understand it. Firstly an IFA is a broker - they offer advice and sell financial products. They literally had to introduce RDR and FAMR a few years back here because the IFA market was unregulated and chock full of cowboys building advice firms just to get a quick management buy out riding on sales commission promises and dragnetting customers at any cost. This control was mostly ineffective, pushing growth by acquisition as the principal method of obtaining capital. And this is even worse in Europe which has somewhat dubious union-wide legislation and local legislation is not enforced by regulatory agencies. Even the standardised T&C coaching process which is a light touch "don't be naughty again" system is actively gamed by most advisers.

In realistic terms the market is a Ferengi death match.

I won't go too far into my credentials in this space publicly but I have over 15 years working in that sector...
« Last Edit: May 01, 2024, 05:33:31 pm by bd139 »

Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo