Yuk

[TomGutman]

When I first saw MC' my immediate reaction was "yuk". After playing with it a bit I have decided that my initial reaction was too kind.

MC' seems to have completely lost its focus. Rather than being a tool to do calculations, with printing provided to allow a record of the calculations, it has become a tool for creating documents (and a very poor one, at that), with some calculating capability thrown in. That's just not the primary purpose of Mathcad.

The whole design process seems to have been based on some really bad principles. As I always say, answers are easy, the hard part is getting the question right. If you ask the wrong question even the right answer will be wrong. Here the design, rather than asking what can be done to make a good and usable calculation package, seems to have set out to answer the question "how can we make this look just like MS Word 2007". A bad question, especially considering the common reaction of users to Word 2007.

If PTC really wants a system that looks like WOrd and creates documents like word, they should start by admitting that and building it as an add-in to word -- similar to the way the MS equation editor is an add in to MS Word. This might be an interesting product in its own right, and a good integration with Word would tolerate a considerable loss of functionality (compared to a real Mathcad) and still be quite useful. BUt such a product should not be confused with Mathcad, or a successor to Mathcad.

The design process also appears to have suffered from the same problem as the so-called beta test. Schedule driven. No concept of doing it right, of determining how much time it takes to do it right. Just an arbitrary schedule into which the process must fit -- whether or not it actually fits.

I'm rather fond of the baby metaphore for programming management. The mythical man-day. The calculation that says that since one woman can produce one baby in nine months all you need to get a baby in one month is to hire nine women to work on it. The schedule driven design is akin to deciding that nine months is too long, doesn't fit the schedule, and so ask what can be produced in five months. We know the answer to that one.

Then there's the integration tab on the ribbon. If ever there was a poor choice of terminology, this is a prime example. While in computer terms integration is quite appropriate, Mathcad is supposed to be mathematically oriented. And mathematically integration has a very common, very well defined meaning. And this ain't it. Actually, the very presence of the tab indicates a failure to have a proper integration. If it were indeed properly inegrated there would be no need for such a tab -- the integration would be seamless and automatic. Variables would simply share a name space, with no additional user input.

And so Ethelred the Unready will ship. Ready or not. Except there's no or, it's clearly not. I PTC going to stick with it's policy of sticking it to the individual user and not allowing them access to any fixes? So that those who do get the product will be forever stuck with all the bugs and limitations in the initial release?
__________________
� � � � Tom Gutman

[ValeryOchkov]

I seems that I know the secret (guessed) of Mathcad Prime.

Mathcad Prime is not a separate software product. It is a future application of MS Office or Word!
Mathcad 15 (Spirit) will be a separate software product!
Val
http://twt.mpei.ac.ru/ochkov/v_ochkov.htm

[PhilipLeitch]

Tom raises some very good points.

HOWEVER - Every software project I've worked on has ended up being exactly what was intended from the outset, even if the customer realised that it isn't what they wanted. I've worked on some silly projects and I spent many hours only to have the customer upset because they got exactly what they asked for.

What's my point? Mathcad Prime that we have just beta tested is exactly what PTC intended it to be. It might not be what we want, it may not be what we will use, but for some reason it's what PTC wants.

PTC developers have actually been very successful and should be commended for a good solid effort.

I for one want to state that I can see the amount of development time and effort that has gone into this product and I think the developers and project leaders should be commended.

Form a development perspective, they have done a top notch job.

But as a product... I totally agree with all of Tom's comments.

Philip
___________________
Nobody can hear you scream in Euclidean space.

[StuartBruff]

On 9/13/2009 4:24:01 AM, pleitch wrote:
== Tom raises some very good points.

As usual.

== == PTC developers have actually been very
successful and should be commended for a good
solid effort.
== I for one want to state that I can see the
amount of development time and effort that has
gone into this product and I think the developers
and project leaders should be commended.

Verification = Good 🙂

== What's my point? Mathcad Prime that we have
just beta tested is exactly what PTC intended it
to be. It might not be what we want, it may not
be what we will use, but for some reason it's
what PTC wants.

Validation = Bad 😞

Stuart

[TomGutman]

I think you missed one of my main points. If you ask the wrong question, even the right answer will be wrong. And the product design here clearly asked the wrong questions. It doesn't really matter whether the programmers implemented the design properly or not. The result is bad.
__________________
� � � � Tom Gutman

[PhilipOakley]

On 9/13/2009 5:25:13 AM, Tom_Gutman wrote:
>I think you missed one of my
>main points. If you ask the
>wrong question, even the right
>answer will be wrong. And the
>product design here clearly
>asked the wrong questions. It
>doesn't really matter whether
>the programmers implemented
>the design properly or not.
>The result is bad.
>__________________
>� � � � Tom Gutman

i.e. doing the wrong things righter!

Philip Oakley

[RichardJ]

On 9/13/2009 5:25:13 AM, Tom_Gutman wrote:
>I think you missed one of my
>main points.

Actually, I think you and Philip are saying the same thing. You start with a basic premise, then go from that to a design, then to the implementation. If the premise is wrong, then what you get is not what is needed, but that is not the fault of the designers or implementers.

Where I do disagree with Philip is that this is what any significant group of customers asked for. I find that very hard to believe.

Richard

[RichardJ]

On 9/13/2009 4:24:01 AM, pleitch wrote:

>HOWEVER - Every software
>project I've worked on has
>ended up being exactly what
>was intended from the outset,
>even if the customer realised
>that it isn't what they
>wanted. I've worked on some
>silly projects and I spent
>many hours only to have the
>customer upset because they
>got exactly what they asked
>for.

And I have worked with software (and hardware) development people that delivered what they thought was asked for, but blew it completely. And when that happens more often then not the developer will never admit they blew it, because if you read the FRS in some really cockamamie way, you could actually come up with what they delivered. The problem is a lack of thought, and a lack of communication. The developer either doesn't think about it enough to realize that what they think is being asked for doesn't make much sense, or just forges ahead anyway, because "that's what was asked for".

On the other hand, I have also worked with customers who didn't get what they wanted, but certainly got exactly what they asked for 🙂

Richard

[PhilipOakley]

On 9/13/2009 9:30:57 AM, rijackson wrote:
> but certainly got exactly what
>they asked for 🙂

>Richard

Very true !

Philip Oakley

[GuyB]

I think we can all agree that, as tested, the product was pretty worthless.

What we're all speculating about, however, is what PTC actually intends to release. I have _such_ a hard time believing that PTC would release such an underperforming product that I can only imagine we were providing feedback on an intermediate step along the path of designing a new interface.

I don't yet have enough experience with PTC to guess at what they'll do from here. I'm hoping that with the feedback they received from the beta test they'll go back and tweak what we suggested as part of the process of re-integrating programming and (I hope) symbolics.

If, however, they release a duct-taped version of what we just saw then I would judge their credibility to be near-zero. Who could possibly expect to have a viable mathematical product with functionality less than[*] a mid/upper range handheld calculator, while coupled to a publishing interface that, frankly, doesn't well lend itself to publishing?

- Guy

[*] The _first_ graphing calculator I looked at online had not only a large library of functions and the ability to graph equations, but also programming and a computer algebra system for symbolics!

http://h10010.www1.hp.com/wwpc/us/en/sm/WF05a/215348-215348-64232-30821-215351-384712.html

[TomGutman]

They've already announced their plans, quite some time ago. Programming and symbolics (whatever they end up with along those lines) are scheduled for the 2.0 release -- at best. The beta test is over, no mention of any further testing. So presumably they take what we have seen, together with a few fixes based on their interpretation of our inputs (since there has bee no feedback on any of the reports, it's hard to tell how much of what was reported was understood by PTC) and whatever fits. The whole design and development process seems to be driven not by what's fit but rather what fits (on the ribbon or in the schedule -- the effect is much the same). They have a schedule for release, set a long time ago, probably in the October-Novenmber time frame, which they are going to stick with. The nature of the product is clearly secondary (or is that tertiary?) to he schedule.

BTW, I would not suggest holding your breath while waiting for the 2.0 features. I don't think they have managed even a preliminary overview design -- just pious hopes. Mona was talking about trying to liberate the ctl-. keystroke from the range definition. While ctl-. never made much sense, and I wouldn't spend much time trying to keep it, the fact that it is an issue indicates that PTC hasn't the faintest idea how the concepts of programming and symbolics are to be implemented.
__________________
� � � � Tom Gutman

[RichardJ]

On 9/14/2009 2:24:10 AM, Tom_Gutman wrote:
> The beta test is
>over, no mention of any
>further testing.

That could be good or bad. If they base what they are going to do next on what we all think, the next beta would not be next week. Or even next month. There is way too much that needs to be done. So maybe they haven't announced it because they don't know when it will be. Or maybe they'll just fix the bugs and release what they've got. I suspect the latter, unfortunately, but one can but hope.

Richard

[IRstuff]

We're the beta testers, implying that there were "alpha" testers. That puts us squarely in the middle of a product release cycle.

I doubt that anyone at PTC is contemplating going back to the drawing board...

Too bad; I just decided to take inventory of my older versions of Mathcad, just in case...

TTFN,
Eden

[RichardJ]

On 9/14/2009 10:32:57 AM, eden_mei wrote:
>I doubt that anyone at PTC is
>contemplating going back to
>the drawing board...

I doubt it too. If you have spent more than two years creating a product, you don't then go to upper management and say "the users don't like it. Can we have another two years to start over?". Not unless you have suicidal tendencies, anyway.

>Too bad; I just decided to
>take inventory of my older
>versions of Mathcad, just in
>case...

Indeed. I still have my copy of 2001. I never had 2001i, which was the last version that did not require activation and is therefore not subject to any control by PTC. I expect I could get a copy if needed though.

Richard

[StuartBruff]

On 9/14/2009 10:40:10 AM, rijackson wrote:
== I doubt it too. If you have spent more than two years creating a product, you don't then go to upper management and say "the users don't like it. ..."

As I've intimated before, without being sure of who it is that PTC are really listening to, it may be a case of "well, some users don't like it, but the majority seem to think it's still a little complicated."

Whilst Prime is very restrictive, it does have a solve block, which can be used to deal with a number of problems that ordinarily require programming. I believe, although I've never owned one, that some of HP's financial calculators operate in a similar manner - lots of useful functions and a good solver, no programming (eg, see http://collab.mathsoft.com/read?127793,103 for some matrix manipulation).

Stuart

[RichardJ]

On 9/14/2009 10:51:33 AM, stuartafbruff wrote:

>As I've intimated before, without being
>sure of who it is that PTC are really
>listening to, it may be a case of "well,
>some users don't like it, but the
>majority seem to think it's still a
>little complicated."

Yes, the big question would indeed be "who are they listening to?". I find it hard to come up with any scenario in which a majority of users would think Prime is too complicated. Or even adequate.

If they are asking existing Mathcad users, how many would say Prime is adequate, let alone better? I think very few, and for sure not a majority. Perhaps they are asking users of other products. I would call these non-users, but maybe PTC would call them users if that other product is Pro/E. In which case it should read: "Most of our current Mathcad users don't like it, but a majority of the handful of Pro/E users we've asked, who have never used Mathcad in their lives and may or may not ever use it in the future, think it's great". I can't believe that's the recipe they used to define the product. So, as you say, who are they listening to? Perhaps the answer is "themselves". Someone or some persons decided that this is what users want, even if the users don't know that themselves (I don't know this to be true, I am just throwing out a possibility).

Richard

[StuartBruff]

On 9/14/2009 11:27:56 AM, rijackson wrote:
== Yes, the big question would indeed be "who are they listening to?".

Indeed.

== I find it hard to come up with any scenario in which a majority of users would think Prime is too complicated. Or even adequate.

Without knowing who the Real Mathcad user base are, or who the intended user base are supposed to be, scenarios are difficult to validate.

Certainly, some of things that have been requested or omitted seem a little bizarre to the type of users who frequent this forum, but these views do hold sway (eg, global variables being sufficiently confusing that they have been deliberately dropped with no apparent replacement for the capability)

== If they are asking existing Mathcad users, how many would say Prime is adequate, let alone better?

A couple of years ago, I was having a telecon with a UK Mathcad rep and mentioned how glad I was that M13 had replaced M13 ... for all the reasons that I won't go into in case it reawakens nightmares. He said that it was only a small number of people who went ballistic at M12 and most users seemed happy. In particular, he mentioned a rather large UK defence research organization with lots of licences, and that picks people for their Brains, hadn't so much as raised a whisper of complaint. Naturally, I fell off my chair in astonishment and still haven't recovered from the shock.

In the end, I concluded that either they'd put M12 back in the box and just hadn't bothered commenting on it (I mean, where to start?) or were hanging off the rafters or catatonically huddled under their desks. However, I still entertain a dark, nagging feeling that maybe they were happy with M12 and could do what they wanted with it; in which case, Spirit may very well be adequate for many users' needs ...

Stuart

[IRstuff]

Indeed, while we do have about 30 licenses for M14 on maintenance, we have yet to install any updates beyond M020, I wouldn't at all be surprised if that company was sticking with M11, or somesuch.


TTFN,
Eden

[PhilipOakley]

I'm still on V11.2a, and attempting to get a stable (now V14 M030) product..

I have 450 users (though many just dabble or read other sheets..)

Do remember that PT have a very large user base of non-mathcad users (draughtsmen & designers) who won't know any different, so will see Prime as different from Excel, and with units! So are likley to be seduced.

Philip Oakley

[RichardJ]

On 9/14/2009 12:01:31 PM, stuartafbruff wrote:
> However, I still
>entertain a dark, nagging feeling that
>maybe they were happy with M12 and could
>do what they wanted with it; in which
>case, Spirit may very well be adequate
>for many users' needs ...

Oh, I could easily believe that a majority of people never used Mathcad at a level where they noticed any difference at all (good or bad) when they switched from 11 to 12. However, comparing the differences between 11 and 12 to those between 14 and Prime is like comparing the differences between a 2005 Toyota Camry and a 2006 Toyota Camry to those between a 2009 Toyota Camry and a bicycle.

Richard

[Fred_Kohlhepp]

On 9/14/2009 12:01:31 PM, stuartafbruff wrote:
>On 9/14/2009 11:27:56 AM, rijackson
>wrote:
>== Yes, the big question would indeed be
>"who are they listening to?".
>
>Indeed.
>
>== I find it hard to come up with any
>scenario in which a majority of users
>would think Prime is too complicated. Or
>even adequate.
>
>Without knowing who the Real Mathcad
>user base are, or who the intended user
>base are supposed to be, scenarios are
>difficult to validate.
>
>Certainly, some of things that have been
>requested or omitted seem a little
>bizarre to the type of users who
>frequent this forum, but these views do
>hold sway (eg, global variables being
>sufficiently confusing that they have
>been deliberately dropped with no
>apparent replacement for the capability)
>
>== If they are asking existing Mathcad
>users, how many would say Prime is
>adequate, let alone better?
>
>A couple of years ago, I was having a
>telecon with a UK Mathcad rep and
>mentioned how glad I was that M13 had
>replaced M13 ... for all the reasons
>that I won't go into in case it
>reawakens nightmares. He said that it
>was only a small number of people who
>went ballistic at M12 and most users
>seemed happy. In particular, he
>mentioned a rather large UK defence
>research organization with lots of
>licences, and that picks people for
>their Brains, hadn't so much as raised a
>whisper of complaint. Naturally, I
>fell off my chair in astonishment and
>still haven't recovered from the shock.
>
>In the end, I concluded that either
>they'd put M12 back in the box and just
>hadn't bothered commenting on it (I
>mean, where to start?) or were hanging
>off the rafters or catatonically huddled
>under their desks. However, I still
>entertain a dark, nagging feeling that
>maybe they were happy with M12 and could
>do what they wanted with it; in which
>case, Spirit may very well be adequate
>for many users' needs ...
>
>Stuart

My company (on the recommendation of several of us) did not install 12. From inertia, they also did not install 13. They are currently checking out licenses for 14, M00--as issued. I'm still running one stand-alone license for 11. The fallout from version 12 still hasn't all settled. The ONLY thing keeping Mathcad alive here is a small but loyal user base and the difference in per seat cost with MatLab.


Fred Kohlhepp
fkohlhepp@sikorsky.com

[IRstuff]

Fred,

How odd; I'm in the same boat. I've got a copy of M11, and the company I work for likewise skipped M12 and M13, but did install M14.


TTFN,
Eden

[IRstuff]

However, we also run Matlab, because there are just some things that run way better in Matlab.

TTFN,
Eden

[GregorySavage]

Ditto. I still work with M11 mostly, skipped M12 and M13, and installed M14 at work. I recall Tom Gutman saying something like "...God intended us to use M11..." in a previous beta test cycle.

I definitely think M' is weak and lacking in spite of some of its novelty -- it should go back to the drawing board and resultant product should exceed the functionality of M14 with all the extension packs. On a side note: The one thing I think is a plus with M' is that it includes the Signal and Image Processing extentions.

I was not so much thinking of the product's integration with Word or Excel (I just want to be able to import Excel data and paste graphs into Word), but I am wondering what PTC might be doing with Mathcad with the intent of integrating it with its other products. I really don't know much about what their other products do, but they seem to be more orientated towards CAD technicians (I am guessing), so a simplified computational tool is OK. I am wondering who PTC is trying to please with M'.

Greg S

[GuyB]

On 9/14/2009 10:51:33 AM, stuartafbruff wrote:
>....
>I believe, although I've never owned one,
>that some of HP's financial calculators
>operate in a similar manner - lots of
>useful functions and a good solver, no
>programming

I've used an HP-15C handheld calculator since 1982 or so - even _it_ has programming. Painful, but possible.

- Guy

Boring anecdote: I remember in high school not believing the claim that if you put 23 randomly-selected people in a room you have a 50% chance that two of them will share the same birthday.

I painstakingly programmed the calculation into my HP-15C and ran it several times before I felt like I really got it.

[GuyB]

On 9/14/2009 10:40:10 AM, rijackson wrote:
>I expect I could get a
>copy if needed though.
>
>Richard

I would expect so ...

- Guy

[SM1]

On 9/13/2009 4:24:01 AM, pleitch wrote:
>
>PTC developers have actually
>been very successful and
>should be commended for a good
>solid effort.

On the contrary, they have been most unsuccessful and done a very poor job. Forget about if this product is an upgrade to MC14 or not. It is very poor implantation of whatever it is trying to do. I doubt there will be takers even if the product is given away free.

>I for one want to state that I
>can see the amount of
>development time and effort
>that has gone into this
>product and I think the
>developers and project leaders
>should be commended.
>

Totally disagree. This product is worthless and I wonder why the developers and the project leaders should be commended. As a matter of fact they need to do soul search.

>Form a development
>perspective, they have done a
>top notch job.
>

It is the most poor job done by professionals. I have seen better implementation of technology by armature programmers.

SM

[PhilipOakley]

Colleagues who have worked with PTC (i.e. Pro/E) say
that PTC buys up companies that it sees a useful
spark in, sucks them up, chews them around, then
they re-appear in a whole new way after having been
worked over...


Philip Oakley

[RichardJ]

I'm not sure what to make of that statement. After all, a piece of meat appears in a whole new way after it has been sucked up, chewed around, and worked over by a dog 🙂

Richard