Any news on a new version?

[itc]

Is it possible to have an update on current progess towards a new version and what issues are being corrected?

[pinbrook]

9.0.2 was released only a couple of weeks ago.

Actinic tend not to announce what will be in future patch releases as it may not be able to fulfill all promises.

Maybe ask a specific question if you have one?

[davie_ne1]

I was told that 9.03 was going to be released shortly due to the problems with the sync server (multi-user systems only effected)

correction re-read tracy's post a revised 9.02 is going to be released shortly to correct sync server probs

[itc]

9.0.2 was released only a couple of weeks ago.

Actinic tend not to announce what will be in future patch releases as it may not be able to fulfill all promises.

Maybe ask a specific question if you have one?

We seem to be having a torrid time with v9 at the moment.

We've stopped using multi-user because we've had a couple of tear your hair out experiences with our db getting corrupt.

Drag and drop for moving items around isn't only causing problems with multi-user it seems as we've had it throw a wobbler in standalone mode (and others have too apparantly).

We have it throwing up php & condition errors (lots of them) in the design on a site we know is fine - exiting actinic and going back in and all is fine again until 5, 6, 10 or 24 hrs later and it happens again.

[simonc]

Drag and drop doesn't work out of the box on single user v9. Drag and drop a product to a different category and it disappears into the ether. Another bug that took less than an hour of use from install to discover.

[AlistairW]

I hate to be overly critical when it comes to 'niggles' and all software when released has problems. However that said, v9 has more than niggles and in my mind the testing process appears to have been rushed.

Features that should have really benefitted users have made the release an unstable bugged disaster. The quality control prior to release is shocking. I hope that in the future a little more effort will be put into testing.

I have never used a more frustrating piece of software - the previous releases had their faults - but at least they worked.

[simonc]

I hate to be overly critical when it comes to 'niggles' and all software when released has problems. However that said, v9 has more than niggles and in my mind the testing process appears to have been rushed.

We were the developers in a similar situation to this with a web app that had an aggressive timescale. We introduced the majority of bugs by attempting to squeeze last minute changes into the release when we should've been doing the (dull but necessary) testing. Absolutely fatal. We've now updated our processes to ban code changes during test periods and so far it's worked.

I have never used a more frustrating piece of software

My feelings are somewhat similar. Coming from an asp/asp.net development background I was willing to accept that a lot of the flexibility I was used to was not going to be available, but that would be compensated for by a mature product that'd allow us to get a site up and running quickly and cheaply. This, sadly, has not been the case, with an unstable IDE that proves remarkably easy to crash by editing the page templates, frustrating editing environment that could really do with being MDI and sometimes allows you to drill down into layouts and sometimes doesn't, a database that required manual recovery after becoming corrupted during a site snapshot (the process that's supposed to protect your work) and numerous minor niggles. Providing some fairly basic extensions (mini cart for example) requires hacking core files which have to be reinstated when service packs are installed and I don't even want to go near that checkout speedup code, much as I would like a smaller number of steps in the checkout. You can't even try these out in a sandboxed site because of the licensing model without resorting to a virtual machine (advantage:immediately restore your munged install to a virgin state).

I expect the Actinic experience gets a bit better the second time around, but I'm not going to be sticking with it to find out.

One final thought: Some of the themes are long past their sell by date, Hi-Tech standing out in particular with its remarkably inaccessible background image. Didn't I see this template in Frontpage 98?

[pinbrook]

One final thought: Some of the themes are long past their sell by date, Hi-Tech standing out in particular with its remarkably inaccessible background image. Didn't I see this template in Frontpage 98?

some themes are well past their sell by date but have to be left in for lecagy reasons (according to actinic)

It has been requested that these old themes get hived off somewhere only to be installed if requested. A brand new v9 customer doesn't need them, but someone upgrading might.

[cdicken]

Is it possible to have an update on current progress towards a new version and what issues are being corrected?

There are two releases to look out for in the coming days and weeks. The first is to deal with the majority of issues experienced by Multi-User customers, and the second is a general update to address performance and stability.

Tomorrow, a new build of 9.0.2 will be released, which should deal with the majority of issues experienced by multi-user customers. The issue was a bug in the Microsoft build environment we use to compile our software, so we've upgraded all our systems over the last few days to a new service pack and rebuilt the software. This is the only change in the 9.0.2 release, there are no other bug fixes.

Then a few weeks after that we'll be releasing 9.0.3. This has been worked on for the last few months to fix the most serious bugs found via internal testing and in the field. It also includes some significant improvements to generation and upload time for large sites (in several of the large customer snapshots we've been using for testing, the upload time was reduced by about two-thirds).

As a general note about the multi-user issues mentioned here, and the comments on testing, we did carry out a lot of testing in our labs on a variety of network configurations, but there are a vast number of different out there, and it looks like a certain combination of network configuration options which cause these issues to manifest themselves weren't included in our test plans. This is why we ran the beta program, but it looks like the people on the beta program didn't have the necessary environmental factors in place either to see this problem. I don't know what else to add really. It would have been much better to find the issue before it was released, but sadly that isn't always possible.

[simonc]

Hmm, they should really be aware that new users are likely to have the same reaction as mine: I couldn't quite believe what I was seeing as I scrolled through that list the first time. First impressions count.

[RuralWeb]

This is why we ran the beta program, but it looks like the people on the beta program didn't have the necessary environmental factors in place either to see this problem. I don't know what else to add really. It would have been much better to find the issue before it was released, but sadly that isn't always possible.

NO- I think that the answer was that people did not really participate in the beta testing as we discussed on my now removed thread. Although its interesing that my comments on there are now in several threads as other people are finding the problems I highlighted.

[cdicken]

Hi Mal - an interesting point.

We had lots of people downloading the beta software, and a good amount of feedback from users like yourself who were testing the software, but I think you are right in that the scale of the testing was smaller than with previous releases. There are probably lots of reasons for this, but suffice to say we'll be enhancing our test programs now for major (and some minor) releases so we are not only relying on internal testing and field testing coming from volunteers downloading the software off their own backs.

[chris ashdown]

I have alsways been suprised that you dont ask a couple of hundred customers to send in their snapshots for your team to try out before upgrades and small upgrades

After all we are all busy running shops online or developing and only have limited time to quickly check things out, and with the last V9 beta we were only allowed single site checks

There is also the problem that many of us do not want to use our ISP servers for testing incase they all get screwed up, so maybe actinic could allocate space on their servers for people to beta check on

[cdicken]

Hi Chris

Testing on our servers is a good idea, but unfortunately it rather defeats the object of testing in a variety of environments in the field.

We did test with about 50 different customer snapshots when were testing upgrades etc (we have a record of the ones sent in to support) but the problem with allowing complete flexibility, is there will always be some element of design we have not accounted for. Flexibility is a two-edged sword.

[pinbrook]

but suffice to say we'll be enhancing our test programs now for major (and some minor) releases so we are not only relying on internal testing and field testing coming from volunteers downloading the software off their own backs.

flippin heck! at last! your commercial developers have been telling you this for years. Especially after v8

relying on volunteer beta testers is suicide, especially after v8. As you can see there are several developers now who have said after all the time we/they gave to v8 we/they are not doing the same with v9, i also note several developers advising potential clients not to upgrade