Moderator note.
This thread has been split off from this thread, as the discussion was blocking real answers.


I understand what you are saying in general, but in this case, I would still be astonished. Doodler is a good example of what I am thinking... How to approach the situation.
I wouldn't consider AutoCAD the same thing, since it is something real, technological, and not something based on dogma.
I used to work with AutoCAD myself (back in the days of version 10, non-windows, Lisp, if that tells you anything)

Don't get me wrong, because I would probably do the same thing on the outside chance that...

__________________
Numbers are not case sensitive. (me)

AutoCAD's about five years from obsolete. Autodesk is pushing their new software called Revit. Byebye independent plans, elevations, sections and details you've gotta update individually. I've got one project in it right now and its pretty amazing. A single 3D computer model that you build in something akin to AutoCAD's model space, then just about everything down to the detail sheets is extracted viewports from the model. All you do in the paperspace view is throw in notes and dimensions, which link to the model, so when the model updates, the notes and dimensions update with it. Section cuts, detail marks, and elevation symbols are automatically linked to a page and update automatically when they move...

It ain't perfect, but its close enough. I still got my grin during training. My first question to the instructor was "Can it do stairs right?" That look of utter defeat in their eyes makes me giggle every time. Someday, maybe, but not yet.

I don't think that Autodesk is going to get rid of the core ACAD package for a little longer than five years, but I am in 100% agreement that they want to phase it out at some point. I have HUGE issues with Autodesk as far as AutoCAD goes, and this is yearly release thing bugs the crap out of me. They really need to get things right over there.

The problem for us is that ACAD does too many things poorly and not enough things well. I do a wide range of CADD, although practically none of it is Architectural. Civil, Geotech, Electrical, Environmental, Surveying, and some Mechanical Engineering -- they all come into play in my line of work. (So does Architectural, when it comes down to that, but there are others around here that are stronger in that field of design than I.)

I find that the newest releases just keep making things worse. Functionality gets lost or goes screwy, and the new things you gain don't always apply to the way you use the software. When we deployed 2007, it took nearly five weeks to sort out all the bugs from the new CUI and the way it hated all of our established protocols for customization, LISP, plotting and X-refs. And it didn't add one single benefit to how we do our jobs, since most of the improvements were in 3D and we design primarily in 2D (at least the engineers do. I work in 3D 'cuz I'm a REAL cadd monkey). Then, about the time we got it all working right, and got coordinated with the satellite offices, ACAD 2008 came out. No service packs for 2007. Just a new release. We're shelving it. Screw it, it's not worth the effort of deployment. Right now, we are simultaneously using five versions of ACAD, two of Autodesk Map, two for Land Desktop, three for Civil3D, and several for SurvCADD, which we use for surveying and civil design. This is mainly because our engineers and designers either don't have the time in their work schedules to train and relearn (Deadlines, people, deadlines!), or they flat-out get tired of spending more time learning the software than they are using it.

In fact, if newer versions of ACAD could communicate better with older drawings, then we would probably still be on 2004. It worked best of all the versions they have come out with since I started CADD work. Many small firms cannot keep up with the pace of training and education that it takes to keep a CADD operator current with yearly releases of new software.

Revit....I hear about it a lot, have always really wondered about that. You make it sound like an answer to your prayers. Civil3D is the civil engineer's version of Revit. It manifests itself on this Earth as Satan himself. There is a push to get this software package to replace Land Desktop, and to wean Autodesk customers off of third-party apps that do civil work that core ACAD can't handle (like generating terrain models, cutting profiles and sections, etc). But, if Autodesk wants to phase out their other software packages in favor of this one, they need to make it work right. Actually, to be more precise, they need to get the AutoCAD core to work right first. I have far fewer problems with the 'extra' stuff than I do with core commands.

I read over this and it came out like a rant, which I guess it kind of is, but there is a point: Do you (and any other ACAD operators reading this) get affected similarly by Autodesk's current businees practices? Do you find yourself having to backpedal as much as you are having to move forward? Are the improvements made to your particular application relevant to the work you use that application for?

We basically stopped progressing in AutoCAD after the 2002/ADT 3.3 release. The jump from that general version based on ACAD 2000 to the 2006 engine was just too huge, it would have required a substantial amount of training to make the transition because the new menus were just too much. I use 2006 in my office, simply for coordination with consultants who use it, I do all the saving back to the 2000 engine version.

Revit, at least the Architectural package, is Desktop on steroids. It works great, even if it does work very differently. Revit is actually NOT an Autodesk product in origin. A smaller developer created Revit, and Autodesk bought them out and made their engine their own. Autodesk is actually only working on the core engine itself, they are not doing any but the most basic library objects which were a part of Revit when it was bought. There is a HUGE movement in users that fileshare models and libraries, and some companies like Kohler are actually making their product's Revit models and handing them out for nada in order to keep their name on the cutting edge.

With Revit, it does answer a lot of prayers, but its not flawless. The utter lack of reverse compatability, and some of its own built in inflexibility tell me Autodesk is about two major generations from having a software I'd call worthy of being "industry standard". It is good, its damned good, but it can be a lot better. I think that will come with time. The fact that Autodesk is leaving the third party market wide open for now is heartening. I think with Revit they might just let it go for the large part, simply because with the level of model customization that can be done at the individual level with reasonably advanced users, they'd never keep a lock on it anyway. Cut the overhead, maintain the engine, and let the end users deal with the bells and whistles. This way, when someone whines about the software not having what they want, they've absolved themselves of the responsibility for adding it.

As far as Autodesk's practices, I remember ages ago, attending a seminar sponsored by CADD Microsystems, our distributor, who told us Autodesk wanted to get away from major releases of AutoCAD. Instead of new engines for the sake of making engines, which gave us junk like AutoCAD 13, they'd do small upgrades to the engine. In practice, this resulted in a new version of AutoCAD every 12 months instead of 18. And yes, the fact that they were still obsolescing softwares more than two generations old was somewhat annoying. Autodesk is to the drafting software industry kinda what MacroShaft is to operating systems. They are ruthlessly aggressive in persuing piracy, which is understandable, considering they're a niche software which comes in right about 3500 USD for a full suite license, and they are the 800 pound gorilla in the business with the lion's share of the market. I won't say they don't deserve a lot of clout, they've put out a killer product, by and large, but they do have some things they're very arrogant about.

Oh yeah, AutoCAD 13: what a disaster!

I don't care what they do re rendering and what not, as long as the datums are clearly defined.

__________________
A person's name, or a mark representing it, as signed personally or by deputy, as in subscribing a letter or other document.

Ouch, I say one sentence and look where it gets me...I didn't think I strayed.

Ok, ok, we got to make the split somewhere, I guess I'll take the responsibility of being the fall guy.

(Said in jest to acknowledge the split...)

__________________
Numbers are not case sensitive. (me)

Nah, my bad. You pegged a subject close to home and I was off and rambling.

Which is probably why it works so well.

I have held this exact philosophy for many years. It used to be up to the end user to customize ACAD to their needs. People who could program in LISP (or nowadays VBA or XML) could, and still can, make a tidy sum of money on the side by writing code for people. One of my coworkers holds that it takes too long to learn a programming language. I told him it took longer to re-up my training every year than it did to learn basic AutoLISP and write my own LISP routines. (Thankfully none of them are very complex.) Plus, LISP is so portable that it can be carried across all versions of ACAD. I use some routines that are more than fifteen years old! Occasionally, upon the release of a new version, you have to go into them and update the name of a command or variable because it's been changed, but that's generally the extent of it.

I think Autodesk tries too hard to make everybody happy; they're trying so hard, in fact, they're ticking off a lot of users. I would gladly switch to another program if I knew it could do what I needed it to do. I'm still with ACAD now really only because it's what everyone else uses, and I have to stay compatible.

Which is about when I was working with it...
Anyway, learning a programming language is not a major factor. It's the logical thinking and the techniques on how you apply it to what you are doing that takes time.
I've written somewhere in the vicinity of 25,000 lines of AutoLisp (long time ago) The language was nothing, it was all about the AutoCad concepts that took the time.

__________________
Numbers are not case sensitive. (me)