Bye Bye AfterShot

[afx]

After reading it, and the other posts on this forum, I think the users them self are spreading the FUD without really having a good reason.

Hmmm,
What is FUD here?
Corel has a bad reputation of buying stuff and mucking it up.
The original crew was fired.
Corel never put any engineering resources into the program, just eye candy and messed up marketing messages (selling it as a cataloging app, which is the worst part of it).
All known facts for those that have been following it closely.
They don't even fix trivialities like the broken D800E support which they claim to support and which could be fixed in a few seconds and a recompile.

I think it is strange reasoning that ASP is some how so complex that it is impossible to learn the code. Especially since Corel has many graphic specialist working for them, and they did have the time to learn the code before the team was fired.

If you do not understand OS kernel code or anything similar, try not to judge this.
This is not your average simple little graphics app.
Nobody said it is impossible, but improbable in a reasonable time frame.
A raw converter is not an app where users are satisfied with one update per year. It is already behind in camera support and will not catch up for quite a while until they manged to understand how this is done. That is what will kill it.

Apart from that, most new functionality can be accomplished with the plugin architecture, which is not that difficult to learn (I'm a developer for 10+ years and I have work on graphics software).

You obviously have not looked at the plugin architecture, nor read what I wrote previously about the new OpenCL support and the plugins.

I think that if they didn't want to support OSX and Linux they could have done that with ASP 1.0. I don't buy the reasoning that they did that to get some cash because that would become a PR nightmare. Apart from that there is no real alternative on Linux so why would they kill the only platform where they don't have any competition.

I tend agree here. This is not a platform specific problem. But it will hurt Linux users most because of the lack of alternatives.

I don't see a reason not to give Corel the benefit of the doubt, and actually I don't see a reason to doubt them to begin with. Esspecialy since they put on there Facebook page that a new version is coming with support for Linux and OS X.

They put out a marketing message. That is all they ever did.
No dates or anything. Words are cheap.

If they miss use our trust we can always gather the pitchforks. Now the only thing it does is scaring potential customers away and is better if allot of people use ASP since it would guaranty continuation of development.

Nobody would complain, if Corel did not have that bad reputation to begin with, and everything we have seen so far just supports that.
A non traceable bug reporting system (you know, serious companies of that size give me tracking numbers etc...), sloppy or no follow up on requests and promises. No interaction with the beta testers, just my way or the highway.
How shall one develop trust in them?

cheers
afx

[stefanve]

After reading it, and the other posts on this forum, I think the users them self are spreading the FUD without really having a good reason.
Hmmm,
What is FUD here?
Corel has a bad reputation of buying stuff and mucking it up.
The original crew was fired.
Corel never put any engineering resources into the program, just eye candy and messed up marketing messages (selling it as a cataloging app, which is the worst part of it).
All known facts for those that have been following it closely.
They don't even fix trivialities like the broken D800E support which they claim to support and which could be fixed in a few seconds and a recompile.

My original question was just that: why should't I trust Corel? I don't know there reputation so please provide more information on why they have this rep.
The FUD is that people are saying Linux support will disappear, and that they will not bring out a new version while there is no reason (other than the firing of the original team, but since we don't know the reason it is hard to say any thing about it)

I think it is strange reasoning that ASP is some how so complex that it is impossible to learn the code. Especially since Corel has many graphic specialist working for them, and they did have the time to learn the code before the team was fired.

If you do not understand OS kernel code or anything similar, try not to judge this.
This is not your average simple little graphics app.
Nobody said it is impossible, but improbable in a reasonable time frame.
A raw converter is not an app where users are satisfied with one update per year. It is already behind in camera support and will not catch up for quite a while until they manged to understand how this is done. That is what will kill it.

[/quote]
Please don't make baseless assumptions that is rude.I'm not saying it's trivial and of course it's is more difficult that your average little graphics app, but then again most things are. But they must have lots of skilled programmers that have knowledge about raw conversion, the basics are more or less the same in every format (except faveon). Also it has very little to do with kernel programming, and have no idea why you would say that.

Apart from that, most new functionality can be accomplished with the plugin architecture, which is not that difficult to learn (I'm a developer for 10+ years and I have work on graphics software).

You obviously have not looked at the plugin architecture, nor read what I wrote previously about the new OpenCL support and the plugins.

I did, I did not, but I'm interested can you point me to that post please

I don't see a reason not to give Corel the benefit of the doubt, and actually I don't see a reason to doubt them to begin with. Esspecialy since they put on there Facebook page that a new version is coming with support for Linux and OS X.

They put out a marketing message. That is all they ever did.
No dates or anything. Words are cheap.

It's only a problem if you don't trust them to begin with.

If they miss use our trust we can always gather the pitchforks. Now the only thing it does is scaring potential customers away and is better if allot of people use ASP since it would guaranty continuation of development.

Nobody would complain, if Corel did not have that bad reputation to begin with, and everything we have seen so far just supports that.
A non traceable bug reporting system (you know, serious companies of that size give me tracking numbers etc...), sloppy or no follow up on requests and promises. No interaction with the beta testers, just my way or the highway.
How shall one develop trust in them?

If that is the case, that would suck indeed, but I'm still curious where the bad rap is coming from

regards,

Stefan

[mick232]

They don't even fix trivialities like the broken D800E support which they claim to support and which could be fixed in a few seconds and a recompile.

That statement is excellent proof that you have no clue about how the software development process works in a software company. Changing the code is a minor effort (it certainly would be in this case) compared to all the activities that have to take place when a new version is going to be released. In fact, just doing what you suggest would be grossly negligent and certainly not in the interest of the customers. You have to go through a full test cycle even after such a minor change and that is just one of the implications.

If you do not understand OS kernel code or anything similar, try not to judge this.
This is not your average simple little graphics app.
Nobody said it is impossible, but improbable in a reasonable time frame.
A raw converter is not an app where users are satisfied with one update per year. It is already behind in camera support and will not catch up for quite a while until they manged to understand how this is done. That is what will kill it.

If you mention kernel code, you better state in what way it adds complexity to ASP because I don't see it. QT is an abstraction layer that is there hide the specifics of the OS. And it is doing that extremely well. Don't overestimate the complexity. Several plug-in developers have managed to dig deep enough into the product to be able to ship all kinds of plug-ins with obviously little internal know-how available to them and no handover being done.

You certainly would have a point about one update per year being a little lazy, but that statement just cannot be made about ASP yet because the product is just a little over half a year old and there has been already one update.

A non traceable bug reporting system (you know, serious companies of that size give me tracking numbers etc...)

So you have an enterprise-level support contract with Corel? Because if you haven't, even if they gave you a bug tracking number as a consumer without contract, nobody could care less about your little bug. Trust me, I know what I am talking about, I work in that area.

[sauron]

That statement is excellent proof that you have no clue about how the software development process works in a software company. Changing the code is a minor effort (it certainly would be in this case) compared to all the activities that have to take place when a new version is going to be released. In fact, just doing what you suggest would be grossly negligent and certainly not in the interest of the customers. You have to go through a full test cycle even after such a minor change and that is just one of the implications.

I can't bring myself to agree with that. If afx is correct about the scope of the required change (and I see no reason to doubt him), then Corel should be able to release a minor update. For an app like ASP, they should have both unit tests and functional tests - if these are lacking then they have a serious engineering problem anyway.

If you mention kernel code, you better state in what way it adds complexity to ASP because I don't see it. QT is an abstraction layer that is there hide the specifics of the OS. And it is doing that extremely well. Don't overestimate the complexity. Several plug-in developers have managed to dig deep enough into the product to be able to ship all kinds of plug-ins with obviously little internal know-how available to them and no handover being done.

QT is a toolkit that provides a GUI agnostic set of UI widgets and object primitives. It is not there to abstract away the entire OS, though is does make porting simpler by abstracting away the interaction with the various windowing systems. The core of a raw converter is the demosaicing algorithms, the various sensor specific work that needs doing, and the numerous graphical conversions that need to operate on full bit-depth data early in the pipeline. Although this is not the same job as an OS kernel, it seems to me that the complexity is comparable - just look at dcraw.c if you doubt that. The plug-in developers have done a fine job, but they have access only to the parts of the imaging pipeline that are exposed by the plugin API. ASP does not wrap an external raw library, so all the core code has to be maintained in house - given this, Corel letting the dev team go cannot help but damage the prospects of the package. Corel will have to get new developers involved, and the only plan that makes sense to me is if they want to integrate the raw processing with PSP, but don't really care about ASP as a standalone product in the longer term. The fact that the only denial so far has been a content free marketing post from Corel does nothing to make me feel any better about this scenario.

We used to see participation from developers in the forums, then that waned and finally vanished behind a wall of silence, which turned out to be the Corel acquisition behind the scenes. There's been no developer participation (with the exception of Jeff's initial ASP announcements) for a long time now, and we now have a trailing edge product with no development team, and no meaningful communication. I'd love to be wrong, but I don't see a happy ending to this story. I'm happy still using ASP for now, but I'm also actively putting some time into moving my workflow to Darktable. On that side of the fence, there are responsive developers, an active community, an open bugtracker, and a vibrant product. It's got a way to go yet, but I have far more confidence in it's future than I do in ASP's.

[afx]

Please don't make baseless assumptions that is rude.I'm not saying it's trivial and of course it's is more difficult that your average little graphics app, but then again most things are. But they must have lots of skilled programmers that have knowledge about raw conversion, the basics are more or less the same in every format (except faveon). Also it has very little to do with kernel programming, and have no idea why you would say that.

I am not making baseless assumptions.
But you state stuff about the plugin architecture that clearly shows you have not really worked with it and you make assumptions about the internals of an app that you don't know anything about and you neglect to read what was written previously about the internals of AS.

I did, I did not, but I'm interested can you point me to that post please

http://forum.corel.com/EN/viewtopic.php ... 47#p254047

They put out a marketing message. That is all they ever did.
No dates or anything. Words are cheap.

It's only a problem if you don't trust them to begin with.

When Corel took over, I was in the same camp, quite optimistic. But their actions so far has shown that those warning about Corel where right.

If that is the case, that would *** indeed, but I'm still curious where the bad rap is coming from

Google.
Check what they did with other apps in the past. Some references can be found in this thread and the first threads when this AS forum was established on this board.

cheers
afx

[afx]

They don't even fix trivialities like the broken D800E support which they claim to support and which could be fixed in a few seconds and a recompile.

That statement is excellent proof that you have no clue about how the software development process works in a software company.

ROTFL.....
Totally wrong tree...

Changing the code is a minor effort (it certainly would be in this case) compared to all the activities that have to take place when a new version is going to be released. In fact, just doing what you suggest would be grossly negligent and certainly not in the interest of the customers. You have to go through a full test cycle even after such a minor change and that is just one of the implications.

Yes, but that again is totally trivial, not a big deal. And any decent development shop will have release procedures adjusted for the severity and impact of a fix.
It is now more than 4 months after the last release where D800E support was claimed.
Surely enough to go through a testing cycle (and, in contrast to you I know how long it takes Corel to do so...)

If you mention kernel code, you better state in what way it adds complexity to ASP because I don't see it.

Why do I have to repeat myself constantly?
I have written about this previously in this thread.
Custom multitasking kernels with custom memory management are not a trivial thing.
Adding OpenCL support to this is not something you do over the weekend.
And then you need to augment the plugin API to work with it.

You certainly would have a point about one update per year being a little lazy, but that statement just cannot be made about ASP yet because the product is just a little over half a year old and there has been already one update.

Quite a while ago and I bet you will not see a significant (with working OpenCL) update within the next months.
But what is more important, I fear that they will not even get new cams supported in the near future.

So you have an enterprise-level support contract with Corel? Because if you haven't, even if they gave you a bug tracking number as a consumer without contract, nobody could care less about your little bug. Trust me, I know what I am talking about, I work in that area.

Hmm, I know plenty of software development companies form tiny to Fortune 500 that will give every customer a bug tracking number.
Corel doesn't even do this with the beta testers, they just don't care.

cheers
afx

[lobo]

Recently I've been searching for the best RAW converter - I have Canon 40D and LMDE. After spending some time I've decided to buy ASP - this seemed the best option at that time. Good image quality, ease of use, speed, Linux support. But now, after reading this - I want my money back, I've filed a case within Corel 30-Day Money-Back Guarantee. When I use paid RAW converter - I want support: frequent updates, bug fixes, new cameras added, etc.

[stefanve]

I am not making baseless assumptions.
But you state stuff about the plugin architecture that clearly shows you have not really worked with it and you make assumptions about the internals of an app that you don't knI ow anything about and you neglect to read what was written previously about the internals of AS.

If that is the case, that would *** indeed, but I'm still curious where the bad rap is coming from

[/quote]


Yes they have there own multi threading and memory management. That statement would be really impressive if it where for the fact that it's the way to go if you create a c/c++ program and thus not really uncommon. I might be mistaking but I looks like you never have worked as a developer your self, Or maybe only at ui or php level so it looks to you as this big scary code base. Any ways this is not the hardest most difficult mind blowing peace of software in the world, people created and learned far bigger and more complex code bases. As for openCL support nobody ever set ASP is easy software that can be build or understand in day's weeks or a view months, but if you have a good code design it would not be that difficult to build it in, yes far more difficult that you average little graphics app bla bla bla, but again there is no reason to think that Corel only employs junior programmers with no skills what so ever.

As for the many companies corel has eaten alive? The only thing I can find is that most big stand alone software products live but with a different name. Yes some stuff they have put together to create an office suite. And a view minor software packages where more like a tech buy out, but I can't find any that they re branded first before killing it with in a year or so. Yes they have re sold some products but that is not killing a product. So if you really have a list of bad behavior just give it instead of pointing to some post in the beginning of the forum. I looked for it but only more FUD hear say post, no facts what so ever. So if you have any please give them as I'm really curious and willing to call out corel but only based on facts

Like state before there is no business reason for corel to kill ASP, nobody has given a reason other than corel is evil "they eat baby's alive really a guy on this forum said so a view post back" no one has give facts only hear say so far.

[jknights]

I think there is a lot of debate but no looking at history.......
History repeats itself.

Wordperfect.... At one time thought to be the best word processor as well as being technically the most advanced. Then Corel bought it...... Ever heard of Wordperfect these days ?
PaintShopPro used to be a direct competitor to Photoshop. In fact it dominated the amateur part of the market in 1990. It has nearly died on its feet compared with Photoshop CS6 functionality. Yes it still exists and it costs less then Adobe Photoshop CS6 but..... Hear much about it now, seen any adverts in photo magazines ?
Same story with other products...... Corel buy and do minimal enhancements and milk it dry!!

Here endeth the first history lesson.

[jknights]

Recently I've been searching for the best RAW converter - I have Canon 40D and LMDE. After spending some time I've decided to buy ASP - this seemed the best option at that time. Good image quality, ease of use, speed, Linux support. But now, after reading this - I want my money back, I've filed a case within Corel 30-Day Money-Back Guarantee. When I use paid RAW converter - I want support: frequent updates, bug fixes, new cameras added, etc.

There is no perfect RAW converter.
I use ASP, Adobe Lightroom/ACR, Capture Pro. Each is better at different types of images.

[stefanve]

I think there is a lot of debate but no looking at history.......
History repeats itself.

Wordperfect.... At one time thought to be the best word processor as well as being technically the most advanced. Then Corel bought it...... Ever heard of Wordperfect these days ?
PaintShopPro used to be a direct competitor to Photoshop. In fact it dominated the amateur part of the market in 1990. It has nearly died on its feet compared with Photoshop CS6 functionality. Yes it still exists and it costs less then Adobe Photoshop CS6 but..... Hear much about it now, seen any adverts in photo magazines ?
Same story with other products...... Corel buy and do minimal enhancements and milk it dry!!

Here endeth the first history lesson.

Wordperfect was king of the hill on DOS but never on Windows, it was first bought by Novell that failed to deliver a good windows experience (partly because of MS), by the time windows 95 and office 95 was released it was already to late. A year later (after win/office 95) they sold it to Corel which really tried to making something of it and even landing some big deals. But with the base Novell and Borland created and the power that MS has it the enterprise space this is and was never a fair fight. So it is really a stretch to say they bought a perfectly good and healthy software suit and let it die (would also be the most bizarre business strategy ever).

As for PSP they bought it half a year before Photoshop CS3 was released, by then photoshop was already the most used (and copied/downloaded) piece of graphics software (I think this was already the case since PS 5, but at least long before Corel bought PSP) . Even the most basic user would rather download PS because that is what the pro's are using and photoshopping was already being used as a verb all over the world. Yes Corels MO is buying companies that are in a loosing battle and than bring some live back in the product and than continues makes money of it. Sure they do not develop at the same pace as Adobe/MS etc but you pay far less for it. I never have used PSP but apparently lots of people are happy users of the product so they must do something right. If you are right that they don't do advertising, they must have good software as they still manage to have a 250 million dollar revenue (2007) company, this can not be done if you stop developing 8 years or older software.

Fact is that both WP and PSP where a shell of there former self and where well bellow the competition in terms of mind share , adoption and capability before Corel bought those companies

[KeithR]

Maybe Corel got fed up with the old development team not fixing KR's issues.

And despite the assertions of the fanboys, they wouldn't be the only ones - they're not just my issues, are they?

it did cater for user expectations long term.

Graham, that's not remotely true - the original team treated Bibble and then ASP as their own personal playground, dicking around with bells and whistles and completely ignoring real, potentially show-stopping usability and image quality issues in favour of whatever gimmicks they were enjoying playing with. One thing you could absolutely rely on with the original team was that faults did not get fixed.

In short they treated their "job" like an open-source hobby project.

Personally, based on my own long experience of Bibble and ASP, this could easily be the best thing to have happened for me as a user in donkey's years: I wouldn't wish unemployment on anyone, but based on my experience of the software, I have absolutely no qualms whatsoever about Corel getting rid of the old team. Maybe now we'll see some professionalism creep into the development and support process and actually see some improvement in the product: God knows there was zero chance of that with the old team calling the shots.

[Blrfl]

Why do I have to repeat myself constantly?
I have written about this previously in this thread.
Custom multitasking kernels with custom memory management are not a trivial thing.

Not relevant, either. ASP makes very good use of threading and may well do its own memory management, but there's no kernel customization involved. If the product required kernel customization to run, they'd sell two, maybe three copies.

I fear...

Fear is the problem. If what you have right now works for you, great. If it doesn't, find something else. If it goes from "works for you" to "doesn't work for you" between releases, you keep using the old release until you find something to replace it.

Hmm, I know plenty of software development companies form tiny to Fortune 500 that will give every customer a bug tracking number. Corel doesn't even do this with the beta testers, they just don't care.

There are just as many that don't. There are also lots of companies that give their customers that kind of attention and still couldn't find their butts with two hands and a flashlight. There's not a whole lot of difference between Corel's black hole and getting a tracking number that doesn't gain you any more information than until the next release comes out and seeing if your bug is still there.

I've done software licensing deals that gave me considerable inside access to the vendor and the product road map, a lot of influence over what went into future versions and the latitude to call the CTO directly and get a custom-cut interim version that solved a problem that needed solving right now. Those deals ran into many millions of dollars, and you're not going to get even a fraction of that kind of a support in a hundred-dollar consumer product.

The company that announced the new product in its pipeline before it was ready for sale and then went bankrupt because sales of their existing product immediately dropped to zero is a classic business school case study. Corel's not going to spill its plans for ASP without a good reason and an NDA, and I don't think I'd commit to a product from a company that did.

In short they treated their "job" like an open-source hobby project.

They had every right to do that. Bibble was their product and the business surrounding it was theirs. When you own it, you get to make the decisions, right or wrong. Without a service contract in place, vendors aren't obligated to provide anything more than a box with a CD in it. Or, as Bob Dylan once said, "just because you like my stuff doesn't mean I owe you anything."

--Mark

[kemp]

Personally, based on my own long experience of Bibble and ASP, this could easily be the best thing to have happened for me as a user in donkey's years: I wouldn't wish unemployment on anyone, but based on my experience of the software, I have absolutely no qualms whatsoever about Corel getting rid of the old team.

100% agree with this. The old team demonstrated they couldn't execute. My hope is that Corel has put in place a new team that can. Only time will tell.

[afx]

Custom multitasking kernels with custom memory management are not a trivial thing.

Not relevant, either. ASP makes very good use of threading

Guess how it does that?
Eric designed his own multitasking kernel because standard frameworks did not perform well enough for his purposes. AS does not use any standard framework for that.

Fear is the problem.

Nope.

If what you have right now works for you, great. If it doesn't, find something else.

That is the problem. AS works quite well enough right now, but not well enough to be perfect.
Anything else on the market neither has a similarly efficient workflow nor covers all operating systems.
And switching to a new tool is a considerable effort.

Corel's not going to spill its plans for ASP without a good reason and an NDA,

Yes. Now think about that for a while and check who of those that where in beta programs did post anything positive....

cheers
afx