Will Komodo 9 finally be the Komodo that has all those niggles and bugs ironed out?

Discussions
1

As anyone who’s been on the forums for the last few years knows, I’ve been a Komodo user since version 5, and am pretty vocal when it comes to the direction of the product. On the flip side, I also contribute extensions, so I like to think it there’s a bit of give and take there.

However, the last couple of major releases have left me wanting more, namely all those “nearly features” made to work like they should do, and just a general tidy up in general. No new features, no new Stackato, no new forum, just the tools we use day-in-day-out working brilliantly.

I’m not alone in my thinking either. The following couple of threads pretty much say the same thing:

Komodo 7.1.0 beta 1 now available
http://support.activestate.com/node/8759#comment-21480

HTML editing in WebStorm vs Komodo
http://community.activestate.com/node/8842

Folks just want awesome QA, and things to work “just right”.

I think this is where Sublime Text has stolen a march on Komodo. It forks Komodo’s CodeIntel, then releases a product that minimizes the bloat - no database editor, regex tools, outliner, or any of the other big-ticket items ActiveState seems to think people want - just the basics, with everything else farmed out to textmate bundles. What it lacks in breadth it makes up for in polish - hence everyone’s talking about it.

Personally, I won’t be switching as I like Komodo too much, but that doesn’t mean there’s nothing to be learnt from Sublime’s approach of making things “sublime”.

I wonder if the same amount of time and effort had been spent on fixing all these gripes as has just been spent on this (fairly needless) transfer to some new forum software, where the product would be now.

Will Komodo 9 be the best Komodo ever?

I really hope so, but I’m not holding my breath.

Cheers,
Dave

1 Like
2

FWIW, my personal gripe-list is here, and @toddw’s seen it… but nothing much seems to have progressed since then (about a year ago) …

3

This is of course our ultimate goal, but one thing keeping us from achieving that at this moment is feedback. We need our community to grow so that it can help us make our product better. So whilst the new forums might seem less important to you in solving your gripes they are in fact created fully in service in resolving said gripes, as well as addressing other needs of the community.

This doesn’t just go for the forums, but for a lot of the things we are doing, the new website, our frequent blogs and screencasts, our participation on github, etc. We’re hoping the community will take notice and grow with us, become a part of the development process. It is vital to the success of Komodo.

All that said, I also want to point out that we are in fact taking a more direct approach to those “minor annoyances” that add up to form one big annoyance on occasion. We fully intend to address much of this in Komodo 9.

4

I’ve been programming for a living for over 30 years so I’ve seen a lot of products for various environments come and go. Since 1999, I’ve been programming primarily in Perl and the rest in C and bash. I’ve been a Komodo customer since it’s earliest days. I’ve been happy with Komodo back in the days I used a PC. Five years ago, I dumped Windows for Mac OS X since all my programming is on Linux and tinkering with the OS is just a waste of my time (I have real work to do).

Anyhow, I’ve seen Komodo IDE evolve. With each new Komodo release since I’ve been on OS X, I get excited because I’m hopeful that performance will improve. It has not. I’m on OS X 10.9, 1GB SSD drive, and 8 GB RAM and nothing else running other than a few SSH terminal sessions (outside of Komodo IDE). I’m firmly in the camp of “do one thing and do it extremely well.” I keep trying to wedge Komodo IDE into my workflow, but just simple text editing is a no-go. I end up waiting for the spinning beach ball to stop so I can continue editing. It’s lousy, to be quite honest.

I agree with Dave Stewart. I suppose the regex tester, DB client, snippet manager, toolbox, etc. are cool extras, but I would give them up in a heartbeat for actual Mac-native performance! It still feels like a Windows app ported to OS X. Bbedit blows Komodo’s doors off in terms of performance and it is old. After each disappointing foray back to Komodo IDE, I return to MacVim which is as fast as lightening. A responsive editor increases productivity, not a collection of whiz-bang extras. Mac users may be in the far minority of Komodo IDE buyers, and that won’t change as long as performance is so bad. I should just cancel my upgrade subscription because the performance never improves.

  • Andy Staab
5

I use Mac pretty much exclusively for Komodo and performance is also a huge priority for me. However we can’t simply focus 100% of our efforts on one single area of Komodo, whilst you guys specifically might be ok with this not all of our users would.

That said, @toddw is spending considerable time and efforts this year to work on optimization and has already made a boatload of improvements. We don’t have a specific date set yet but we do expect to be making the Komodo 9 builds available “soon” for everyone to help test as we continue development on it.

6

Thanks for responding, Nathan. I understand the sacrifices when targeting three OSs with one code base. I’ll give V9 a try and let you know how it performs. Thanks, again.

7

Hi Nathan,

I took a few days to think about your reply, but I can’t say I agree with this at all.

The product has to come before the community, or else there won’t be a product worth having a community around. But I understand that you’re saying is that the forum is just one component of what looks like a “re-framing” exercise for Komodo, at least in marketing terms, which is no bad thing - money in the coffers and all.

Back to my main point of feedback though - there really is no need for further feedback on many of Komodo’s outstanding issues. The forums are full of gripes, moans and examples from the last 3 Komodo versions, and the bug tracker is there to formalise them into a coherent stream.

My main worry for the dev team is that there’s just so many little niggly rough edges (and I mean that in the most respectful-possible way) that it must be hard to know which ones to tackle first - though I suspect this is a management issue. And what might seem simple on the face of things, may often have knock-on effects for another component or subsystem.

I would love it if there was some kind of concerted effort to get through them all, perhaps a “Bug-fix Friday” (and for all I know, there is) where the most impactful small-scale niggles were lined up against a wall and shot, relentlessly, week-on-week, until they were all gone. Focus on the most effective bang-for-buck ones first (menu item ordering, breakpoint indicator in non-debugged languages, quotes around HTML attributes, Replace Dialog pinning not working, etc) and move up the scale of PITA as the smaller ones are pushed aside.

Fingers crossed this is what your more “direct approach” is about!

Cheers,
Dave

8

I agree with you that the app itself is more important than any other aspect of the project. However I don’t think it is in any sort of dire state that we cannot afford to focus on other aspects. Yes there are bugs but this is true for any software project. I agree that there is a pile up of small bugs that can add up to make your experience frustrating if you are unlucky enough to bump into several of them and I want to assure you that we are very much aware of them. The way we generally work is we work on big features first and once those are done we start work on smaller features and stability / bug fixes. I intend to focus particularly on those small, old bug that on their own are not a big issue but altogether end up being one big issue.

This IS on our roadmap and whilst I cannot guarantee a bugfree release of Komodo I can tell you that it is our explicit intend to deliver a release which addresses these “small annoyances” more so than we have in previous releases.

9

Hey Nathan,

Thanks for the replies, and reassurance.

I have the KE repo downloaded at my end too, so if I can be of any assistance in some of these tasks, I will certainly make myself available for that.

All the best,
Dave

10

I’m really hopeful too. Komodo reminds me of Perl 6 in some ways (but maybe that’s unfair)…

Sadly, I must concur with Dave’s assessment… I’ve read so many forum posts where a user complains about something not working, etc, and he/she will receive an initial response suggesting a fix/workaround, which often does not resolve the issue, and then… silence… queue sound of crickets.

I think I last tried Komodo 7 or something, and it lacked some basic autocomplete feature/something, so I returned to vim (I don’t have time to spend ages figuring out how to fix the tool). Now, I’m back (sucker!), and it’s v8.x, and there are still basic issues.

It would be so great if things just worked as expected, and at an acceptable performance level. I’d pay for that. But I cannot pay for something which does not do the basic productivity stuff right.

I want to believe, I want to embrace, but, just, can’t.

11

Of course this can happen from time to time but I don’t think it’s fair to say that this happens most of the time, if it does happen that we inavertedly lost focus of a thread you can always bump the thread. What does seem to happen often is that people simply dont answer when we ask them for some type of elaboration on their problem, either because they’ve moved on or solved the issue but didn’t let us know about it.

Also note that the forums are meant for community support, we hang around to help people since our forum community is not big enough yet to sustain itself but we hope to get to the point where our interference here will be minimal such that we can focus on developing Komodo. If you want reliable and official support I’d suggest purchasing a support subscription.

12

Komodo guys , you guys gotta be HURRY!.
I really love Komodo but Atom.io is taking over and it is very fast. (Development Pace , commit rate is Scary and as of 0.124 now , it is quite stable).
Komodo should be A LOT MORE Hackable than Atom.io if it done right (XUL +Python is a lot more powerful than coffeescript + less) . Main thing is Atom.io is very very easy to build and Extreamly easy to hack. Within 1 weekend , i had hacked alot of things , modified a few plugins , modified themes , modified Atom code to fix a bug (EWW I am allergic to Coffee and now Allergic to Coffeescript too , so disgusting piece of shit language.)

If you guys can simplify build process , MANY Of us will swarm in , and Contribute you.

13

We’re not exactly competing with Atom, given that they are an editor and we develop an IDE. That said we ARE simplifying a lot of the development process, but its currently not heavily documented yet as it is all a work in progress. Documentation will follow as we near the Komodo 9 release, but keep an eye on our wiki - https://github.com/Komodo/KomodoEdit/wiki

Links of interest:

14

A simplified build process would be nice - I tried a week or so ago to compile and, well, just gave up. To say it was onerous would be an understatement :smiley:

Thank heavens for http://downloads.activestate.com/Komodo/nightly/

Pity about the mozilla dependency… what a monster, holy crap.

15

Please be sure to check out the Docker build steps in case you are on Linux, this will make building significantly easier.

16

Docker works fine only on Ubuntu, huh.

17

No it works fine on any linux distro that supports docker. What makes you think it only works on Ubuntu?

18

It’s doesn’t work on Debian 7, huh. It’s just download a lot of packages and nothing else. Anyway you can compile Komodo if you install dev-package of mozilla.

19

Well you still need to run commands on it, its not just a matter of starting the Docker container and thats it.

Once your container is started you can run Komodo with util/docker/docklet run

20

Development process need to be simplfied a lot.
For runtime , komodo should deliver patched , prebuilt binaries , it takes a lot of time doing that (And failing at my end) .
I have been contributing in atom.io alot due to very easy build process.

We’re not exactly competing with Atom, given that they are an editor and we develop an IDE.

its no longer true as current version of Atom.io is very capable of an ide, at this rate of development it will be full fledge IDE ,in 6 months time. Its plugin system is very easy to modify and develop . Everything of Atom.io is extremely easy to modify and extend. Comparing amount of contributors to Atom vs Komodo is like Comparing Rain forest against a Desert. When i switch to Atom, Only missing are intelligent auto-completions. Other features are doing much better than komodoedit already.

I love to contribute in Komodo because it is written in my favorite language , Python and I still love Komodoedit more , can we step up the game pls?