[JKT]

Originally Posted by Ilgaz

I know they are Windows issues but how can we get true info/true critism/true security check of OS X version?

WebKit is entirely OSS. There isn't anything to hide.

[zottel]

Originally Posted by troyb

We hope to get a beta of OmniWeb 5.6 out soon. Sorry for the wait everyone.

Is there anything like at least a vague timeframe you could give us? I'm not the annoying kind of user, normally, but the JS bug has been around for just too long now; the current version hasn't been updated for quite some time ... All in all, the feeling of using OW has fallen from "top notch" to "nice features, but outdated".

What about sneaky peek versions like pre5.5? Nobody expected perfect stability or all features working then, nobody would expect these now. It was fun to take part in testing and reporting bugs, and everyone was satisfied even though not everything was working—we could see the progress in every new version we downloaded, and that's the point. It's frustrating not to know what's going on and having to cope with a browser that's great in its features but not really stable and limited in terms of compatibility.

Just my 5ct.

[JKT]

What's to say that there aren't sneaky peeks of 5.6? Regardless of whether there are or not, it is self-evident that OmniWeb has been relegated to being "not of much importance" at OmniGroup since OmniPlan and OmniFocus have become the darlings of their development cycle. It is now over a year since the first 5.5 sneaky peek was out, and the JS bug still remains unfixed. That is exceptionally poor treatment of your customers, by anyone's book.

[zottel]

So you're even more angry than I am. :-) Or rather :-(

It took some time till the JS bug was fixed in WebKit, but still this was rather long ago, so it's more than about time that we get a working OW.

Personally, I'd be satisfied if at least I could use some beta versions of 5.6 and really see that there's work being done. But all the "wait for 5.6", "we hope (!) to be releasing 5.6 soon", "sorry for the wait" stuff has really been going on for far too long now.

[Chiller]

Originally Posted by JKT

What's to say that there aren't sneaky peeks of 5.6? Regardless of whether there are or not, it is self-evident that OmniWeb has been relegated to being "not of much importance" at OmniGroup since OmniPlan and OmniFocus have become the darlings of their development cycle. It is now over a year since the first 5.5 sneaky peek was out, and the JS bug still remains unfixed. That is exceptionally poor treatment of your customers, by anyone's book.

I agree as well. It can't be a matter of communication because the Omni Group blog is update fairly often. With the exception of Safari, just about all other browsers have their status published on a regular basis or the work is visible to see from others.

Omni Group. Look, it's like this. I really like your browser. I am a licensed user of 3 of your products. I would like to respectfully request a little more detail than soon. 2Q? 3Q? 4Q?

[mshalperin]

Unfortunately, OW never developed much market share in the non-Safari OS X market - possibly because it's the only non-free product from a company looking for sales profits. The developers of the free browsers obtain their revenue and support by other means. The Omni Group lowered the price but this hasn't had much impact. They probably decided to shift their development resources into more profitable products. I suspect that if it isn't sold to another developer, it may be soon dead in the water. In contrast, Shira, which is free, released v2.1 on 6/19 "to support new Web Kit, installed by Safari 3.0."

[Jon Hicks]

I was also hoping that the move to webkit would mean less hassle for the developers, but it seems as though they still have to spend a long time adding their customisations to WebKit in order to retain those Omniweb features. Time between releases is just as long as it always was.

I guess what I would like to know is not so much 'when will 5.6 beta come out?' more 'is there a life for Omniweb after 5.6?'. Are Omnigroup still committed to developing it?

[Ken Case]

Sorry if things have seemed quiet on our end: we're just as anxious to get 5.6 out there as everyone else, and we've been putting continuous effort into 5.6 since before we even shipped 5.5.

We adopted the Leopard version of WebKit in our 5.5.x releases, which was great because it was much faster, included a lot more web features, and fixed some Tiger WebKit bugs. Unfortunately, midway through the Leopard WebKit cycle Apple introduced a very major architectural change to WebKit, where they reimplemented forms to no longer use native Mac OS X widgets. Reimplementing all those widgets from scratch had some big upsides for web compatibility (and probably for the iPhone as well)—but these new untested widgets introduced hundreds of bugs in the way web form widgets work, so we decided we couldn't ship something based on that work until they got through those bugs. We branched 5.6 to start tracking Apple's latest work (which we tried to merge with our code base every week or two) while simultaneously trying to cherry-pick some fixes which could also help 5.5 and apply them back to the our 5.5.x releases (which, remember, were based on a work-in-progress Leopard WebKit). As time went on, though, more and more of the new fixes couldn't be applied to 5.5 (the code was too intertwined with their new forms implementation)—but we felt that the version of WebKit underlying 5.6 still wasn't ready to go into a shipping release.

Thus, the apparent limbo of the last few months.

We haven't just been sitting on our hands, of course: not only have we continued to track the latest WebKit development in OmniWeb 5.6, we've still had a lot of work to do ourselves to reimplement our enhanced forms on top of the new non-native widget base (features such as zoomed editing, text completion, and saved state in Workspaces), and just to continue testing with each new integration of WebKit to make sure our other features (such as ad blocking) don't break as they continue rearchitecting the framework.

Most of that work was just a one-time thing, though, reimplementing our features on top of the new form widgets. We then hit a period where integrating with the latest changes to WebKit would generally take us just a few hours—but we were still waiting for WebKit itself to get stable enough to ship something to you.

Finally, a ray of hope appeared a few months ago, when Apple decided that they would start focusing their WebKit effort on stabilizing the features they already had (such as fixing those hundreds of new form widget bugs) rather than continuing to introduce new features (destabilizing things even more), and since then they've fixed most of the forms bugs that we thought it would be unreasonable to include in a shipping, supported browser. (Apparently they think so, too, since they just shipped a beta release of Safari 3.0.)

I've had both OmniWeb 5.6 and OmniWeb 5.5.4 on my dock for months, trying to spend most of my time in 5.6 but having to switch back to 5.5.4 regularly as I encountered problems in the new WebKit. But it's been getting better each month, and it's now been a few weeks since I've had to launch 5.5.4—which seems like a good indicator that we have something we could share with all of you. At this point there are just a few more regressions from 5.5 which we're trying to fix:

• <bug://bugs/35924> (Regression: From third-party sites ad block setting doesn't work)
• <bug://bugs/38732> (Regression: "Load All Images" doesn't seem to work)
• <bug://bugs/37286> (Regression: WorkspacesLoadFromCache no longer works)
• <bug://bugs/37431> (Regression: Workspace scroll position is not restored on launch / reload)
• <bug://bugs/35921> (Regression: Form Data isn't restored between workspace launches)

As you can see, this list is getting pretty short and most of these issues aren't so bad to live with (at least, not relative to the pain of using 5.6 through the last nine months). In fact, the most important thing to me on that list is third-party site ad blocking, but that setting seems to be working for me so I'm not sure why it's still listed. (Maybe there's some detail in the implementation that we want to redo?)

Again, I'm sorry that things have seemed quiet on our end, and that 5.6 has taken so long! We originally thought that Leopard would be shipping by now, so we didn't imagine it would take nearly this long for its version of WebKit to reach a stable state. But the good news is that we think we're very close to shipping a 5.6 beta (perhaps with some of the above issues listed as known issues).

[philonous]

Wow. Thanks for the detailed reply. You know that a lot of people are looking forward to the next beta. Don't forget how many people participated in the sneaky peek debugging last year. So, if that program was helpful, it may be fruitful to bring it back.

Payam

[Ken Case]

Originally Posted by Jon Hicks

I was also hoping that the move to webkit would mean less hassle for the developers, but it seems as though they still have to spend a long time adding their customisations to WebKit in order to retain those Omniweb features. Time between releases is just as long as it always was.

Oh, and just to address this point specifically: now that we're on WebKit, it really has gotten much easier to integrate our changes with Apple's. The big delay in 5.6 has primarily been waiting for WebKit itself to be stable enough to release again.

I guess what I would like to know is not so much 'when will 5.6 beta come out?' more 'is there a life for Omniweb after 5.6?'. Are Omnigroup still committed to developing it?

Yes, we definitely are. Nothing else comes close to the user experience and feature set I expect from my primary web browser.