Hacker News Reader

Comment by iterati | Hacker News Reader

This is sadly true. I switched off Chrome 5+ years ago due to the "vim-like" plugins being much better on FF than Chrome (specifically, my keybindings not working on built-in pages) and since FF56, vimperator and pentadactly no longer function. There have been months of back-and-forth between some developers and the FF team, but the functionality required for these plugins has never been prioritized and I had to disable automatic updating and download FF54 so that everything would function. Today I just gave up caring and installed cVim on Vivaldi (https://vivaldi.com/) which is Good Enough for now.

View on HN

Give it some time. I was expecting not to switch because of lack of Tree Style Tabs until around end of year. But I switched some weeks ago. Some addon authors use this opportunity to make a better addon. A few days ago GreaseMonkey has been released with a new API. And of course there will be new addons made by other people to replace lost functionality. It's happening, just not overnight.

That's the issue - there's no real path to making these legacy plugins work. There are open issues for a keyboard api for FF that has had very little movement over months, so while I'm sure it'll eventually get to a point where I can use FF as I used to, that not being the case now has given me little reason to stay with FF and so back to Chrome I went.

It's too soon to judge, I think. I would be happy using the old FF if the new one didn't have already the things I need. In the mean time I would try to collaborate. Once FF 57 is out of beta officially, there will eventually be enough people with itches to scratch. It was in my plans to implement Tree Style Tabs properly if no alternative appeared before end of the year.

It's not too soon to judge - plugins I used no longer work unless I turn off auto-update or move to the LTS version. The team knew these were breaking changes for months or years, but the APIs to un-break things don't exist and aren't prioritized. It's discouraging to collaboration when you look at 2+ year old issues (https://bugzilla.mozilla.org/show_bug.cgi?id=1215061) that still haven't been properly addressed.

You make it sound like there's no movement, but I see enough movement to have both short-term alternatives, and medium term fixes. Read comment 46 and the next ones. They were super busy with what they can actually release with FF 57. Until today the issue wasn't really pressing. In a few days we'll see if they're serious about that issue.

You don't break something and then prioritize fixing it. That's backwards. If they knew it was going to break the proper thing to do it prioritize the fixes before it effects end users, not after. If I were to push up a breaking change this morning and just wait for bug reports, I'd be looking for a new job by this afternoon.

I would agree with you normally. But in this case there's a much bigger "bug" they must handle: The vast majority of users have migrated to Chrome or some other browser that don't respect their privacy.

We power users are like 1-2% of all FF users. I think it's acceptable to break some things for a while to bring back much more users than that.

In the mean time, they've lost one of those power users (likely more, going by some of the discussion in issues for various vim-like plugins) and will, in the future, need to do something to entice me back. This experience has left a bad taste in my mouth and I'm not going to be rushing to go back to a browser that breaks functionality on me without prompt.

They lost me once for terrible, terrible performance. For a year or two until I really needed tree style tabs back. They need to prioritize. Deal with it.

Also, since you're a power user, why don't you use that proposed API already? It seems you don't even need to compile the browser or anything, just enabling a flag and installing the API as an addon.

> They lost me once for terrible, terrible performance. For a year or two until I really needed tree style tabs back. They need to prioritize. Deal with it.

I am dealing with it. I moved to a different browser.

> Also, since you're a power user, why don't you use that proposed API already? It seems you don't even need to compile the browser or anything, just enabling a flag and installing the API as an addon.

Simply not true.

Search for Tridactyl, which is aiming to reimplement the Vimperator/Pentadactyl feature set in WebExtensions Firefox. It's still in early stages of development but there is already an initial release on AMO with some basic features implemented and with a very familiar feel.

I've tried Tridactyl, but the plugin not working on Firefox pages is the same frustration I had with vim plugins for Chrome.

If you were heavily utilizing the nmemonics provided by those vim-like plugins (I was before the upgrade) you may be quite pleased by the fast minimal experience bought by qute-browser which is basically webkit wrapped in python/qt bringing a browser built centrally around vim interaction (as opposed to a plugin)

If the vim stuff was the only plugin I needed, I'd move over in a heartbeat.

Saka keys on Firefox has been a suitable replacement for me.

Not me personally, but there will always be a vocal minority who will complain about having their workflow broken. Any replacement will be missing some critical feature. For a person affected by this, it is a painful experience having their entire workflow broken so they resort to dire warnings on forums such as this one ('Firefox is finished without <my niche plugin>').

In this case, I think Firefox made a good choice in modernising the codebase and the extension APIs. In 2 years we'll find out if the vim people were right or they were indulging in hyperbole.

Less "dire warnings on forums" and more "slow development pace for required pieces to make my workflow even possible again".

Here's an issue opened June 9 that hasn't been addressed by the Mozilla team and hasn't had any updates in a month: https://github.com/cmcaine/keyboard-api/issues/1

Here's the request for a better keyboard API (referenced in the above) that's been open for over 2 years with no end in sight: https://bugzilla.mozilla.org/show_bug.cgi?id=1215061

I'm sure the new FF is going to make some users happy, but as you said, having your workflow broken is quite painful, and the Mozilla team has known about this for years but still not made the upgrade path bearable.

> with no end in sight

Except an already developed patch and proposal that needs some time from the main devs to be reviewed. For a long time they were very busy making FF bearable for 90% of the users. Now that it's released we'll see if this issue is as stuck as you imply.

I guess we will, but with the slow pace that any of this was (not) done over the last number of years, I'm not holding my breath.

eh... I use it now, and it's OK. It's missing a ton of features that Vimperator had, and it will likely never get them.

>specifically, my keybindings not working on built-in pages

This is a restriction for chrome-store extensions. You can install one manually without with "feature"

Reply on news.ycombinator.com