I want to be completely clear, I don't know how this will be received, I don't know if this is the right place, I don't know if an actual person was behind the specific tweet that directed me here. I do know the my experiences are genuine, and I believe that they could very well be widespread.
I don’t think there’s much for me, as a user, to discuss with the community on this matter. I used Mathcad 14 (unless the engineering department is even cheaper than that), and I had a terrible, terrible experience with it. I am told (I have not bothered to verify) that Prime is a substantial improvement over 14. If so, then whenever someone like me uses 14, instead of Prime, at the behest of their professor, that is a lost sale. Not because such people aren’t buying the latest version at that time (it’s the university buying site licenses, anyway), but because they won’t buy any version. Ever.
So, students being taught Mathcad 14 are a threat to PTC’s bottom line due to terrible (but apparently incrementally less terrible than the previous version?) UX. The obvious course of action is to encourage universities to upgrade, or, failing that, simply stop teaching 14.
The way I see it, any further discussion would center around forming an initiative, to come up with a strategy, work out promotions, do risk analysis, budget for losses, maybe set up a social campaign or two. I won’t discuss any of that without a salary.
Actually, that’s not quite right. There is one, potential, point, of contention. Perhaps my experience isn’t universal, and people like me represent a small portion of the marketplace, who’ll pick at anything without being satisfied. Aside from the fact that a discussion on the Mathcad discussion site couldn’t possibly resolve that, how about I reiterate a few of my complaints?
(For the purposes of this paragraph “Mathcad” refers to Mathcad 14.) At a base level, I’m not sure that Mathcad is even suited for the kind of calculations we’re learning with it, over in UMass ChE. That aside… Mathcad is a wysiwyg formula editor, which means that it’s difficult to replicate a symbol on a print-out if you don’t already know what it is. Mathcad, by default, recalculates on update, which wouldn’t be a problem… if calculations didn’t block input and have no obvious way to send an interrupt. I have never, ever seen an error message that gave me more informational content than “something went wrong”, which is terrible if you’re trying to figure out what went wrong. I encountered many of these error messages attempting to replicate precisely a sheet given to me by a professor, which more often than not resulted in a complete breakdown of the system. According to the professors trying to get us to use it, Mathcad will occasionally develop invisible chunks of formulae that can only be gotten rid of by recreating the sheet and hoping it doesn’t happen again. *checks tweets from last May* OH YEAH. It also tries to autocorrect my parentheses, or something. I forget the precise details, and I honestly prefer it that way. Also, sometimes the curves on graphs outright refuse to print. My professors loved it when I told them that.
In conclusion, I want to say, I’m sorry I’m being a bad user, but it’s a nigh-inevitable consequence of being forced (by the department, I should be clear) to use a bad product.
Sorry if this is in the wrong place, but it seemed about right, and I figured I'd listen the twitter for once.
Max,
Thanks for taking the time to write. You offer up quite a bit, so let me try to unpack it and address the larger, key issues.
-Alan Belniak (a former Mathcad product manager)
I'll try to respond to everything in turn.
Max;
I don't know you, you don't know me. If some of my comments offend you let me apologise in advance. But your rant begs for a response.
From what you've written, you're a student of Chemical Engineering. To me, this says that you haven't been exposed to some of the complex software systems that comprise the engineering tools of today. Most of them require a significant dedicated effort to learn, many of them are career software--you can be a NASTRAN engineer (you run NASTRAN.) Compared with those (and PTC owns a few) Mathcad has a much less daunting learning curve. (My opinion, clearly not yours.)
I use Mathcad 14.0 (the original buggy version) at work. (My company has a full site license but our IT group hasn''t bothered to post the upgrades or version 15 yet.) I have version 15 at home. The differences are small but aggravating. I also have Prime 1 at home and have seen Prime 2. Whoever told you that Prime was a substantial improvement over 14 or 15 either has a vested interest or is smoking something; by PTC's own statements, Prime will be missing features that exist in 15 (and 14, 13, 11 . . .) for at least two more iterations. Not that you (or I) care, but Prime will integrate better with some of PTC's other software.
As for the "terrible operating interface (UI, UX?)"--As I said above, every complex software package is difficult to learn. (Mathcad had a major editor change about fifteen years ago (version 7?) and we all cursed steadily while learning the new one. Prime is another such major shift, and the complaints are just as strong. (Do you really like the ribbon?) But more disturbing is the fact that PTC is selling software that's missing major features.
I read your post, and I hear that learning this is hard, it should be easier. You want easy? Go find a pad, a pencil, and a sliderule. (Well okay, a calculator.) That's about as easy as it's going to get.
I will agree that PTC is missing outon the educational front. Most of the young engineersin my company, when presented with a problem too big for EXCEL, reach for MATLAB. Why? Because MATLAB is almost free in school, so they learned it, that's what they know. Now I hate programming. I've done FORTRAN and BASIC, and I've tried to learn C; and I'll concede that MATLAB is easier than those. But I can read a mathcad (wysiwyg) sheet and it reads like a math book, and I can follow what's going on. It's a lot more work to unravel a MATLAB script. Ther's not much that MATLAB does that Mathcad can't.
Max, I reject completely your claim that Mathcad is a "bad product."
To be honest, it might be that part of the reason I was able to write the rant at all was because I've been able to grasp enough of Mathcad to figure out what I have a problem with. There's so much less for me to say about the truly awful stuff they've tried teaching us.
In my opinion, what I find easier to grapple with is something like a mathematics package for Python. We also had to learn MATLAB, which, coming from years of on and off messing around in Python, just feels kind of strange. (My professors do nothing to help the matter when they deliberately indent things funny. The IDE indents for all of us; why would they take that away? Anyway, that's for another time and another place.) Where Python shines for me in that it provides a stack trace, so I can follow what's going wrong with a calculation; it's modular, so I can, for example, swap out different physical models without duplicating or destroying work (this might exist in Mathcad, but I never saw a good way of doing it); it's easy to place a conceptually simple but algorithmically complicated operation inside a function call, and related to that; it's scoped, so, if I'm doing a multi-step calculation with many temporary-use variables, I can name them whatever I want without accidentally obliterating some long-running bit of information; and one final bit is just something we'll have to disagree on. I distrust wysiwyg immensely. I like being able to put names to things when I calculate. (For the record, if I'm doing formatting-intensive writing on my own, I use LaTeX. Formatting-light, I use a plain text editor, and I only use wysiwyg word processors in collaborative work.)
I hope that covered everything; I feel like I skimmed this post somewhat.
Max,
It's pretty clear to me that your mind is made up. It's also clear that you're comfortable programming. So let me just comment on a few of your concerns and I'll wish you well; in the real world any legal way you get the answer is right, so we each get to choose our own tools:
Comment 1:
"Where Python shines for me in that it provides a stack trace, so I can follow what's going wrong with a calculation:
Any time you want an intermediate answer in Mathcad, you move to the right and type "variable name =" and (if automatic calculation hasn't been disabled) Mathcad willto give you the current value of that variable.
Comment 2:
"it's modular, so I can, for example, swap out different physical models without duplicating or destroying work (this might exist in Mathcad, but I never saw a good way of doing it); it's easy to place a conceptually simple but algorithmically complicated operation inside a function call."
I'm not exactly sure what you mean by "different physical models," but disabling one function and inserting a different function with the same name sounds like it might do this. (You DO know therre are functions, don't you?
Comment 3:
" I distrust wysiwyg immensely"
Too bad! I prefer to read what looks like a pencil/scratchpad sheet rather than decipher computer code; that may be the greatest difference between you and me.
I'm done! If you have a method that works for you, great. Go for it.
I like Mathcad, that's what I'm going to use.
Max,
there is a very interesting fable - The Fox and the Grapes
See http://en.wikisource.org/wiki/The_Fox_and_the_Grapes
There was a Time when a Fox would have ventur’d as far for a Bunch of Grapes as for a Shoulder of Mutton; and it was a Fox of those Days, and that Palate, that stood gaping under a Vine, and licking his Lips at a most delicious Cluster of Grapes that he had spy’d out there; he fetch’d a hundred and a hundred Leaps at it, till at last, when he was as weary as a Dog, and found that there was no Good to be done; Hang ‘em (says he) they are as sour as Crabs; and so away he went, turning off the Disappointment with a Jest.
THE MORAL OF THE TWO FABLES ABOVE. ‘Tis Matter of Skill and Address, when a Man cannot honestly compass what he would be at, to appear easy and indifferent upon all Repulses and Disappointments.
Mmm, always a danger. For the record, I have succeeded in using Mathcad sometimes, but I feel like it's been in spite of it.
Hi Max,
I love Mathcad, but I also can appreciate some of your frustration. I think it is possible that you may have been introduced to Mathcad in a "non ideal" manner.
I, like most users I suspect, learned Mathcad "on my own"...no courses, no training, and I didn't go very frequently to the "help desk" of the original company. The key I think was that I started slowly, learning the interface with small, algebraic problems. I was not in school, but I was working in my career. I purchased my first copy on my own, but when it became obvious that it was one of my major tools at work, I was able to get my company to purchase updates. Upon retirement, I have been paying again for my updates.
Back to the learning process, I found the tutorials and help manual to be the main way to learn new features. I am sometimes amazed at the requests for help that show up on this site when just a close examination of those resources would show the user how to do the task.
In your discussion, you had some specific complaints that can be easily fixed. First, you mentioned that you can't input code while Mathcad is computing and there was no interrupt. There is..just click Esc. Sometimes it takes awhile, but it does stop the calculation. An even better approach is to turn off the automatic calculation feature. I don't have MC 14, but in MC 15 it can be turned off by "tools>calculate" on the toolbar.
I may be wrong, but you may have been exposed to advanced capabilites in an advanced course too early in the learning process. And there may have been a time pressure to get something done too fast. I had the same problem when I was exposed to a hybrid analog/digital computer. I had used digital computers but not analog. The professor said, do this, this, then this, and this. and that's all there is too it. Then he left the room. I never did get an analog problem solved. Fortunately, we could use just the digital computer for the course if we wanted. In your case, you may never use Mathcad again, but I hope you do give it another try.
I would like to see Mathcad introduced first in the computer lab course, or even better, in the department's AIChE Student Chapter. Often, those chapters have the upperclass students tutor or lead workshops for the underclass students. This would be a non credit, low pressure way to learn the tool that would then be used in later courses.
Finally, you are right about the error messages not being very informative. But at least they show an error. Many of the errors that they show, and prevent further calculation, would not even be caught in other programs. They would just go on merrily calculating and getting a garbage answer.
I have to agree with other comments, Prime is not a complete tool yet, but it may be what you need to learn. By being limited, it may slow down the learning process. However, there is the danger that you learn it and then find that you can't do something that you need, but it could be done in Mathcad 15. I don't remember MC 14 being that bad, but it may have had some bugs, like most big programs.
I hope you follow Alan's recommendations and follow what's going on in the Chemical Engineering space on this site. It might encourage you to give Mathcad another try. My worksheets are posted in Mathcad 15 build M005. If you can't read them, let us know by posting a reply here or starting another discussion in the ChE space. I could attach a .pdf version of the worksheet that you could read. My hope is that by showing some good ChE examples, people like you will be encouraged to try the program.
Thanks for telling your experience. Sorry it was not good, but your comments may help improve the product or the training and university marketing program. I'm not a PTC employee, just someone that wants to see Mathcad succeed because I think it's good for my consulting business, and I hate to see others using what I regard as inferior tools.
Hi Max,
I am the Senior Academic Program Manager for Mathcad. In my work I interact with many students and professors regarding their experiences with Mathcad. We take feedback such as yours very seriously and distill it in our planning for revisions to Mathcad.
In fact, one way of looking at our development of Mathcad Prime is to view it as an effort to develop a product with an intuitive feel for new users. Your generation of students has grown up with very high expectations for a product interface -- you expect to be able to use it out of the box. I think that we have succeeded with the interface of Mathcad Prime 1.0. In your case, we may not have reached the level of functionality necessary for your current course.
Mathcad Prime 2.0 will be a major release for us -- we think it has all of the capabilities necessary for undergraduate programs to make the transition from Mathcad 15.0 to Mathcad Prime 2.0. I hope that your program will provide Mathcad Prime 2.0 on campus next semester.
In the meantime, feel free to reach out to me directly for assistance.
Chris
Mr. Hartmann;
I would dearly love to me a freshly graduated engineer that knew Mathcad. Within the last three years my group has hired five new engineers and I've been pushing Mathcad at them as an analysis tool. (I've been successful because as group "senior" engineer I can push this issue; but they have begun to see the inherent advantages that I see: the ability to track units and unit balance, the fact that the boss, even not knowing Mathcad, can read and understand the mathematics without having it translated, etc).
Of the five one had heard of Mathcad. Of the five, three were proficient in MATLAB. When I showed them that we could do what MATLAB could and still have the advantages of Mathcad, I began to make serious converts.
The cost comparison between a MATLAB "seat" and a Mathcad "seat" has our sooftware budget manager pushing Mathcad. Three years ago his seat usage for Mathcad went from one (me) to just about six; but it hasn't grown since. The last info I have (last year) put the MATLAB "seat" count at 20+, with people complaining that we needed more.
You want to sell more Mathcad to industry? Then GIVE IT AWAY TO TEACHERS AND STUDENTS. You need to push this program out to the schools;teaching the new engineers about it won't sell it, teaching the engineering students about it will.
Hi Fred,
Thanks for your comments. It is great to have a champion like you to mentor young engineers. PTC is working hard to encourage young people to consider careeers in science and engineering and we use Mathcad to promote their interest and skill. Here are links to two of our flagship programs:
FIRST Robotics
http://www.ptc.com/company/community/first/
Real World Design Challenge
http://www.ptc.com/company/community/education/schools/students/real-world-design-challenge/
All participants in each of these programs can receive a free copy of Mathcad Prime to support their participation in the program and further their academic achievement in school.
We are also working hard to get the word out to colleges and universities about Mathcad Prime 2.0. We believe that it is an outstanding product that all engineering students should learn to use during their first year in the program. I order to support this initiative, we sell the full commercial version of Mathcad to schools as a Mathcad Academic Edition for a very reasonable price.
If we can reach out to any colleges in your area and let them know that companies are looking for students with Mathcad expertise, please let us know. Our user base is one of our strongest assets!
Warm regards,
Chris
You want to sell more Mathcad to industry? Then GIVE IT AWAY TO TEACHERS AND STUDENTS. You need to push this program out to the schools;teaching the new engineers about it won't sell it, teaching the engineering students about it will.
Good point Fred. This has been covered many times in the forum with collabs indicating that Mathcad needs to be pushed into University's. Mathcad wasn't used in my University in the UK either.
Mike