Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

In case the Flow guys are reading this, please please please put out a simple HTML version of your thing.

Flow sits at the top of my "most annoying web app ever" list, because it follows the pattern of loading an empty shell, then populating it. I bet this works amazingly well for the guys developing it from their local dev server, but for those of us across the Atlantic from your servers it's really really painful.

Workflow involves clicking a link to look at a task. Then hitting the back button. Then waiting several seconds for the list to slowly repopulate itself. Then trying to scroll down (which is no easy feat since there's no scroll bar), and just before you get back to where you wanted to be, the list jumps back to the top. Repeat (scroll/jump/scroll/jump) several times before the page settles down, usually cutting the last item in half with no way to see it.

So yeah, Step One: HTML in a table. So that the back button works, and the page otherwise behaves like a web page.

Step two: Un-reinvent the scrollbar. The javascript thing you have in place that only works with the mousewheel just plain doesn't work.

Glad to see you're doing well for yourselves, but I'm sorry to report that I'm one of the people in your "churn" column.



If you have feedback, why don't you email them directly?

Edit: Ok, I was downvoted for my comment. The reason I posted this is because the parent wrote a long comment about the product that has nothing to do with the content of the article and may not even be viewed by the team.


I'm guessing you don't actually use flow day to day. I'm also guessing that if they made an HTML version you probably wouldn't use that either; you probably have your own way of managing your projects and todos. I'm also guessing you're using this post as a sounding board for your gripes with single page javascript applications. Are any of these guesses accurate?


No. I used Flow on a daily basis for the better part of two months, to coordinate a team of six on a major product feature we were developing. The things I mention above are chronic issues I ran into every day, not superficial things I saw once before dismissing the thing.

They're also 100% fixable.

If you're going to do this sort of "empty container" app, it's worth modifying your approach a tiny bit and instead going with "filled container". That is, sending the actual page content over the wire, then doing all your smartness and auto-updating behind the scenes where the user can't see it. You get all the benefits that the browser gives you for free such as cache management and back-button support, and you can also build in all your rich UX just like before.

And of course, if you're going to rip out a 20-year-tested native control like the scrollbar, it's up to you to make absolutely certain that the thing you replace it with is better. Theirs is demonstrably worse in pretty much every way a scrollbar can be worse, and actually fails the simple case of "scroll the document", as it tends to lose its position. The fix for this is as obvious as it is easy to implement: use the native scrollbar to scroll your list.

If they do those things, they may win me and my team back.


Boy was I wrong. Thanks for the clarification, these are great critiques.


You sound like the type of customer I'd be thrilled to lose.


Just because he doesn't use it any more doesn't mean the feedback is any less valuable.

It's great feedback. It points out specific problems and solutions. Ways that, from his point of view, the UI could be improved.


What I'm questioning is whether or not he actually ever attempted to use it as his main project management solution in the first place, and instead gave a knee-jerk reaction to the fact that it was a single-page javascript application. Someone who's asking for a plain HTML version of a rich desktop style application probably wasn't going to use it in the first place. I'm open to eating crow, though.


It's pretty clear from his comment that he used it.

> Someone who's asking for a plain HTML version of a rich desktop style application probably wasn't going to use it in the first place.

That's a very unfortunate comment. In the end the technology is what matters the least in his feedback - odds are if you make the RIA responsive enough, use smart caching, sensible navigation, respect default behaviors, etc, nobody will ask for a "plain HTML" version. It's about the experience and usability.


My guess is you sound like a troll.




Consider applying for YC's Fall 2026 batch! Applications are open till July 27.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: