No programming? Really?

[GuyB]

Is anyone else stunned by the lack of programming in MP1.0?

Am I experiencing a 1:40am lack of cognitive ability, or is there really no intent to add programming to the release version?

I could not envision using Mathcad on a regular basis without the programming ability.

To test that thought I randomly selected ~15 of 243 Mathcad files in my catch-all Mathcad folder (populated by more one-off sheets than any of my other work) and found only 6 sheets bare of the programming construct.

And of those 6, 3 used symbolic math.

Thus, even in a directory chosen to have the highest rate of "whiteboard" sheets I only came up with a 40% rate of non-programming sheets.

Assuming symbolic math is gone, too, then 80% of the functionality I reach for in Mathcad would vanish in this paradigm.

- Guy

[IRstuff]

On 8/14/2009 1:56:04 AM, GuyBeadie wrote:
>Is anyone else stunned by the
>lack of programming in MP1.0?
>
>Am I experiencing a 1:40am
>lack of cognitive ability, or
>is there really no intent to
>add programming to the release
>version?
>
>I don't think I could possibly
>envision using Mathcad on a
>regular basis without the
>programming ability.
>
>To test that thought I
>randomly selected ~15 of 243
>Mathcad files in my catch-all
>Mathcad folder (populated by
>more one-off sheets than any
>of my other work) and found
>only 6 sheets bare of the
>programming construct.
>
>And of those 6, 3 used
>symbolic math.
>
>Thus, even in a directory
>chosen to have the highest
>rate of "whiteboard" sheets I
>only came up with a 40% rate
>of non-programming sheets.
>
>Assuming symbolic math is
>gone, too, then 80% of the
>functionality I reach for in
>Mathcad would vanish in this
>paradigm.
>
> - Guy
>

Those are listed, along with a bunch of others as not being in 1.0, but it's unclear when it will be available. Bear in mind the even E-books, components, 3D plots, etc. are not in 1.0. However, since the install files have Mupad folder, the loss is, presumably, temporary.

TTFN,
Eden

[IRstuff]

Oh, I missed the wording at the top of the file:

Some features will not be developed until Mathcad Prime 2.0. Some other features are not yet in the Beta software or are known to be broken.

TTFN,
Eden

[GuyB]

On 8/14/2009 2:18:36 AM, eden_mei wrote:
>Oh, I missed the wording at
>the top of the file:

As I read it in the table at:

http://www.ptc.com/products/mathcad/beta-test/known-issues.htm

"Symbolic Math" and "Programming, user-defined dlls, scripts" are "Not in Mathcad Prime 1.0"

That doesn't seem to leave much room for interpretation - it looks like they're scrapping those components of Mathcad, at least for 1.0. And there is no assurance that they'll be back for later versions.

"Some features will not be developed until Mathcad Prime 2.0" does not give me a warm fuzzy with regards to programming - I don't see it as a 'feature' so much as "that's what makes Mathcad work."

I need programs to define functions which take one set of inputs but require multiple steps to process. PTC cannot possibly invent & compile all the multi-instruction functions I'd want.

How can you imagine operating on real data sets without the ability to loop and check conditions on the data??

- Guy

[ptc-1178444]

On 8/14/2009 2:52:20 AM, GuyBeadie wrote:
>"Symbolic Math" and "Programming,
>user-defined dlls, scripts" are "Not in
>Mathcad Prime 1.0"

Yes, that's a non-starter for what we do with Mathcad at my company. We can find workarounds for scripts and symbolic math (Maxima for example), but user dlls are key for the data we process. So I guess we're not part of the target customer base for this product.

Nevertheless, the absence of any programing capability has me very perplexed for a math program...

Xavier

[TomGutman]

I might have been had I not been told of that before downloading. Programming is supposed to be in a later release (probably 2.0), symbolics is more of a guess (what are they going to use for a symbolic engine?). Mathcad prime is in no way a replacement for the existin Mathcad. PTC seems to have recognized that, as they intend to continue fixing MC14 even after MCP is released (not that that means much, with a "support" policy that keeps fixes away from ordinary users).

Of greater concern is what has been kept of MC. I've just scratched the surface, but already I've noted:

two dimensional arrays with no proper distinction between matrices and vector

1e307 for ∞ (instead of a proper infinity)

Math styles (now called labels) with no reasonable way to apply them. Made worse by a structure that seems to often require the application of styles.


The need for proper array handling (absolute minimum of three subscripts, and that is very minimal) has been recognized for a long time, and was accepted as a requirement (long term) by Mathsoft. That PTC could do a complete rewrite and not fix arrays boggles the mind, and bodes ill for the future development of Mathcad.

I can better understand the omission of programming. That is a major feature and requires a lot of work to implement properly (or even workably). The bean counters want a product on a schedule (doesn't matter if the product is actually useful, bean counters have no concept of utility, and a good sales force can sell ice boxes to eskimos), so that part of the implementation is deferred. But there is no excuse for such a poor array handling -- arrays are implemented, it doesn't take much extra effort to do it right rather than copying the worst of Mathcad's features.
__________________
� � � � Tom Gutman

[TomGutman]

I got distracted and forgot to complete the post.

At least they got rid of SUC. And even the older restriction on mixed unit arrays. That is good. But that one feature does not really justify a brand new implementation, with the loss of most functionality.
__________________
� � � � Tom Gutman

[mzeftel]

Improved programming and symbolics are top features for Mathcad Prime 2.0. And Mathcad 14, unlike previous releases, is not going away until Mathcad Prime is fully featured.

Labels are not intended to take the place of math styles. Math styles and text styles did not fit in the schedule. Our main efforts were on replacing the math equation editor.

Mona

[TomGutman]

>>Labels are not intended to take the place of math styles.<<

Could have fooled me. Labels look to me remarkably like dumbed down math styles. Dumbed down as there are now a smaller number of them, with fixed names, and no ability to specify the font face or size.

Another area where MC' just takes the bad from Mathcad and strips it down is results formatting. Not only do we still not have the ability to request a fixed number of significant digits (the normal scientific/engineering requirement, fixed number of decimal places is the usually accounting requirement) but can't even control the automatic invocation of scientific notation based on range.
__________________
� � � � Tom Gutman

[mzeftel]

You can format labels under the Formatting tab.

Mona

[TomGutman]

Yes, there is a section of the ribbon under formatting that applies to labels. But it's very limited, dumbed down from the MC math styles. You can apply bold, italics, and underscore, and also choose from a limited palette of colors. No provision (that I can see) for specifying the font face or size.
__________________
� � � � Tom Gutman