[c-smile]

Except of few ideological mistakes, I would say, quite a lot of time was wasted on doubtful technical decisions ...

One year was wasted on trying to make it as WPF application, .NET all that ... 5, if not more, years were wasted on making it work as Electron'ish like application (CEF was integrated).

Evernote is a typical companion application - the thing that sits in tray and pops up as quickly as possible when needed.

Management read once that "Premature Optimization Is the Root of All Evil" but missed the critical point - this kind of application MUST be lightweight or dead.

The elephant was born instead.

[mjburgess]

> Premature Optimization Is the Root of All Evil

Means don't drop down to asm to save a few cycles. It doesnt mean "dont fill your 10fps GUI app with crap".

Today, it more or less should be killed-off as a phrase. There isn't anyone optimizing anything, let alone "prematurely"

[pradn]

> There isn't anyone optimizing anything, let alone "prematurely"

Yes, but in a slightly different way than one might think at initial read of this. Most of the performance choices for a user-facing application have been made by the time the language and UI framework are decided. And these are just chosen out of inertia, to conform to a company standard or to use what you already know.

There's little one can do to make a WPF or CEF application super responsive.

[mjburgess]

I'd be interested in seeing if there's a YT video out there of a WPF expert doing just that.

It might be a matter of hacking-in a caching layer or something, but it feels like probably more things are possible than people think.

A framework is just code, and you can pick-and-choose.

[c-smile]

It could be interpreted that way too.

But usually managers interpret it this way: let's mold it now from that pile of shit and firewood and think about concrete foundation later. VC driven development they say.

It makes sense in some cases of course. But not in cases when concrete foundation is the cornerstone feature of the application. That kind of application shall serve users during whole their lifetime. And so application core shall not rely on particular OS and its features.

[tenpies]

> Except of few ideological mistakes, I would say, quite a lot of time was wasted on doubtful technical decisions ...

Not just technical, but remember when they expanded into physical products that made no sense? I don't mean the handful of scanner implementations - those made sense even if they were made obsolete. The notebooks with "smart" scanning were okay at the time as well, even if they quickly became obsolete too

But I mean stuff like backpacks and socks.[1] That never made sense to me.

I kind of wish Apple had never made iPod Socks, because then we could have a very good indicator of sock-selling as a sign that things will not end well.

Other socks I have around: Snap, Yak, a handful of defunct delivery apps.

---

[1] https://twitter.com/mlevchin/status/281668296628129792