bogs wrote: Sat Dec 02, 2017 5:14 pm
MaxV wrote: Sat Dec 02, 2017 3:09 am
but when a new number goes out, no other old version should come out.
I don't think I'd agree with that statement at all, after all, there is not (or wasn't at the time of this statement) a stable 9.x version. Also, I think for a goodly number of reasons fixes should be backported as far as they will go (without causing new issues).
What Max described is actually the work plan. But if you manage projects of this complexity you'll appreciate the deviations that sometimes occur with any plan.
As planned, there would only be two versions being worked on at any given time: bug fixes added to the most recent Stable release, and those bug fixes merged with features in the latest Developer Preview release.
But once in a while there's a need to introduce a new feature sooner than waiting for the rest of a DP scope to be completed. Moments like that are what gave us the current 8.x sub-branches. The v9 DP branch is pretty awesome with new features, but that scope means it will be a while yet before it's Stable.
So if the company or a key partner or a significant portion of the community needs a particular feature ASAP, we see another version in between the last major (left side of the version's decimal) Stable and the next major release. This allows Stable to remain stable, keeps large-scope work moving forward, while allowing an occasional new feature to land in something in between for those who need it.
It's not ideal. But with any complex project in any industry, little is.
However, Max's statement about not having the download section so completely full is valid, and I remember saying that before.
Oh yes, definitely. But a poor UX is not "vaporware", nor does it mean LiveCode is "dead". By that definition LinkedIn would be "dead" (OMG worst UX I've seen from a company their size, but that's another story).
Identifying issues with the user experience is productive.
Communicating those with anyone other than those who can act on the observations is less productive.
Muddying the conversation by introducing irrelevant alarmist language is counter-productive.
Max frequently has a lot of great ideas. I would encourage him to have confidence in those ideas - they're often quite good on their own, and don't need the excitement of provocative language to drive their point home.
His observation on the Downloads page is, IMO, spot-on.
In fact, I felt so strongly about it I did the thing that seems oddly uncommon here: when I had an observation I wanted to share with them, I shared it with them. I didn't write you, or my wife, or anyone else not in a position to act on it. And it turns out that by writing to them, it became possible for them to write back.
By writing to them I've learned they're fully aware that the current "Wall of a Thousand Links" (a pet term I've come to use for the Downloads page) is not a universal solution.
It's great for experienced devs who may need access to any version made in the last decade.
But of course it's not a great UX for newcomers, triggering confusion and overwhelming anxiety if they encounter it.
And that's why they don't.
And this is the role of research when advocating UX improvements:
Those of us who've been using LC for years know about the Downloads page and visit it often. And because we're so familiar with it, we often share the URL with others looking for a particular version.
But that's not how newcomers get LiveCode.
Good UX practice requires us to set aside our own personal work habits and try as much as we can to imagine the flow of things to the new user. Here that means clearing our cache and our cookies, and starting where newcomers start, at the front page.
When we take the time to start at the front page and look for a download option in the way that a newcomer might, we'd find that we'd have to work very hard to find the Wall of a Thousand Links page. Most will probably never come across it at all.
Much more readily accessible is a Download button, near the upper-right when call-to-action elements are most commonly found.
When the user clicks that button, if they're a new user they're asked to fill in a brief form so a direct download link is emailed. For existing customers, the form action immediately proceeds to download the most recent Stable build.
No one ever sees the Wall of a Thousand Links except us old-timers (and those we keep sending there <g>).
IMO there are still opportunities for further refining the initial download, particularly with the .org Community site about which I have no shortage of opinions.
But since none of the other users here in this user-to-user support forum have control over that server, I'll continue sending my suggestions directly to the company, and would encourage the folks here to do the same.
If you do, please use your time and thiers wisely: take a moment to double-check that the user experience you're writing about is indeed the same as the newcomers you're advocating for.
now 
