« first day (335 days earlier)      last day (4155 days later) » 

12:48 AM
0
Q: Low quality posts

m_goldbergThe low quality posts review queue has become available to me. As I write this post the queue is empty and there is nothing to serve as an examples of what kind post gets put into it. So I ask: what qualifies a post for relegation to this queue? I ask because there a couple or recent posts that I...

 
 
1 hour later…
1:49 AM
Maybe someone can come up with a hack to have auto-completion with templates for user-defined functions. Take a look here:
I think the completion information is stored here: FileNameJoin[{$InstallationDirectory, "SystemFiles/Kernel/TextResources/English/FunctionInformation2.m"}]. You will need to Uncompress the file after importing it as String. The actual usage messages are also in that directory, stored in FunctionInformation.m. If you look at most packages, the ::usage messages are plain text only, but when you load the package, and ? the symbol, you'll get a nicely formatted usage message with italicized arguments. This comes from FunctionInformation.m, not from the package file. — Szabolcs 46 secs ago
 
 
2 hours later…
3:31 AM
@Matariki i think we initally just need to hear what the publisher says.
 
True. As Todd has the connection will you ask him to convey our ideas or do you intent to approach McGraw-Hill
 
I am assuming he will read the comment stream and take action then maybe report back.
 
Well, never assume... I'll monitor the thread for a while and will see what happens. Nothing will happen anyway so short before xmas.
 
 
4 hours later…
7:26 AM
@Mr.Wizard after this answer, I have to wonder: what do you do for a living? I probably should have asked a long time ago, but I admit, I'm curious.
 
8:17 AM
@rcollyer Why, because it's obvious I have no idea what I'm doing and it took me four tries to get a proper answer? :-p
 
 
2 hours later…
10:44 AM
@Gabriel Why do you say this? CDF is a really great concept, and at least with Player Pro one can do cool things.
 
11:42 AM
@Szabolcs Thanks for the troubleshooting. Yes, I only have the newest version, downloaded yesterday, and none other installed under any Mathematica directory (base, user, instalation). I've searched for SE Uploader.nb under these dirs, and only the one installed was found. BTW, I'm still using Mathematica 8.0.4.
@Szabolcs
1) Yes, the problem appeared when I've deleted the old version of the palette and installed the new
2) I both installed it manually (empty palette appeared after full Mathematica restart) and via the built-in *Install...* menu after removing the previous installation (ended up with the same result).
3) {TaggingRules -> {"SEUploaderLastUpdateCheck" -> 3.5647520510711589*10^9, "SEUploaderVersion" -> "110"}}
@Szabolcs The only version of the palette is C:\Users\Z\AppData\Roaming\Mathematica\SystemFiles\FrontEnd\Palettes\SE Uploader.nb, 137,570 bytes. It appears as: i.imgur.com/IHME9.png
 
 
1 hour later…
12:54 PM
@IstvánZachar Got the same problem here.
@IstvánZachar Can you try to start Mathematica with the -clean option?
 
1:30 PM
@halirutan I've tried it: frontend was recached/rebuilt, but result is the same empty palette.
 
@IstvánZachar are you on Win?
 
@halirutan Yes, Win 7 64-bit, Mathematica is 64-bit as well
 
Can you look inside the file $UserAddOnsDirectory/FrontEnd/init.m? There should be a section SE Uploader..
Something like:
PalettesMenuSettings->{
 "SEUploader.nb" -> {
   WindowMargins -> {{Automatic, 447}, {Automatic, 24}},
    TaggingRules -> {"ImageUploadHistory" -> {}}}
@IstvánZachar Unfortunately, when I had this issue, I didn't resolved it. I just tried various stuff and then it worked.
 
2:18 PM
@Mr.Wizard That's not what I was thinking. But, four different methods does indicate you have more than a passing familiarity with programming in general, and likely a good bit of higher mathematics, too. So, just curious.
 
2:41 PM
@halirutan There is a piece of code like that in the init.m, and the history even holds my last image I've uploaded successfully (with the previous version).
 
2:59 PM
@IstvánZachar So what you can try is: Backup the init.m and delete this part:
"SEUploader.nb" -> {
WindowMargins -> {{Automatic, 447}, {Automatic, 24}},
TaggingRules -> {"ImageUploadHistory" -> {}}}
And this part
TaggingRules->{
"SEUploaderLastUpdateCheck" -> 3.564623772483001`16.304588688240703*^9,
"SEUploaderVersion" -> "110"}
It's still possible that there are some cache values, but at least this is all I could find by searching my whole .Mathematica directory.
@IstvánZachar Additionally, you could backup the palette and open it with a text-editor.
Try to delete at the beginning all cache which is wrapped in Mathematica-comment and looks like
(* Content-type: application/vnd.wolfram.mathematica *)

(*** Wolfram Notebook File ***)
(* http://www.wolfram.com/nb *)

(* CreatedBy='Mathematica 9.0' *)

(*CacheID: 234*)
(* Internal cache information:
NotebookFileLineBreakTest
NotebookFileLineBreakTest
NotebookDataPosition[ 157, 7]
NotebookDataLength[ 25959, 530]
NotebookOptionsPosition[ 25626, 514]
NotebookOutlinePosition[ 26049, 532]
CellTagsIndexPosition[ 26006, 529]
WindowTitle->SE Uploader
And at the very end, you do the same.
@IstvánZachar If this really doesn't help, I'm afraid we have to debug this more carefully. But then, someone needs at first to reproduce this again.
 
 
1 hour later…
4:27 PM
@István @halirutan That palette is not so complicated ... it is just so annoying that these problems keep cropping up and I cannot reproduce them.
@halirutan It's possible to prevent the front end from including the outline cache. There's an option, see the option inspector. "FileOutlineCache". There's also the cell change times, which are better disabled for version control.
@halirutan I manually removed both the cache and the front end version info (which is a Notebook option) from the palette notebook. I did this to prevent the v8 front end from throwing warnings if I happen to generate the palette with the v9 front end.
 
I noticed most bugs don't have a version tag, and figured I should go through and ad them, but I'd feel like an ass bumping everything tagged bug to the top, any way to sneak a tag edit in without bumping?
 
9
Q: What to do with [bugs] questions now that version 9 is released?

Szabolcs@MikeHoneychurch asked in chat what we should do with questions that have been tagged as [bugs] now that version 9 was released. Some of these bugs might be fixed in version 9. This post is to start a discussion for deciding this.

This is also relevant
 
@Szabolcs Yes, and this is a good idea! The only annoying thing is, that I cannot develop easily.. Every time I build the uploader notebook, git let merge this when I do a git pull..
 
@Szabolcs how did you get such a fancy inlined link?
 
4:37 PM
@ssch Just paste the link, it'll inline automatically.
@ssch It's called oneboxing. It works with SE sites, Wikipedia, SE blogs, Youtube, etc.
 
@ssch You have to paste the link as separate message!
 
ah, need to be only that line
 
@Szabolcs Do you really think it is possible to write a (template)expansion function which works in the frontend?
In that case: I wrote a small camel hump expansion demo which works like the devil!
 
Hm? Template expansion?
Do you mean the Ctrl-Shift-K thing?
 
@Szabolcs Yes, you mentioned it here:
@Szabolcs here..
 
4:43 PM
FunctionInformation2.m, you mean?
The bird is going crazy, keeps screaming in the cage. It just wants to come out. But I can't even type something today without it trying to remove the keys from my keyboard (or the piano keys when my wife is playing piano) ...
 
yes... I maybe show it better. I'll make a small screencast of what I've tried.
@Szabolcs How good is your wife with the piano?
Can she play something like that here:
It's currently my favorite.
 
She has small hands, unfortunately.
Which reminds me of:
I'd have to ask her if she can play that.
@halirutan The problem is that I am really not sure how FunctionInformation or FunctionInfromation2 are loaded. Maybe it's through FrontEndResource.
 
5:00 PM
This is hilarious..
 
Where can I find that function that saves a notebook as a .png on imgur?
 
@ssch It's the uploader and you can find it here:
55
A: Can I easily post images to this site directly from Mathematica? (Yes!)

SzabolcsThis is my implementation (with contributions from @halirutan and help from a number of people) of an image uploader palette, which I would like to share with the community to make it more convenient to use this site. Download the latest version here! INSTALLATION: Download the palette note...

 
@halirutan I don't think he means that
14
Q: For Sale! Three potentially useful palettes

belisariusI wanted to learn how to make palettes, so I've made this bunch. They are still beta, but I want to share them for cross-platform checking and bug hunting. The functions aren't mine, and proper credit is included. I've made very few modifications to the original author's code in order to fi...

Third palette
 
@rm-rf that's the one, thanks
 
5:16 PM
@rm-rf ups.. OK.
 
I voted to close this as too localized.
 
@Szabolcs "while with w_ it did not look so very bad..." reminds me of the girls in physics where the formula was used which fits the best ;-)
 
Exactly ;)
 
@Szabolcs madness >.<
In Mathematica : SetOptions[EvaluationNotebook[],FrontEndVersion->Null];
In bash:

$ grep -i frontendversion /tmp/test5_nfev.nb
RowBox[{"FrontEndVersion", "\[Rule]", "Null"}]}], "]"}]], "Input",
FrontEndVersion->"8.0 for Linux x86 (64-bit) (October 10, 2011)",
 
5:31 PM
@ssch I know!
It seems it's always written by the Front End, so NotebookSave seems to be out of the question.
 
Yea, $Version didn't like to get tinkered with one bit :D
 
I was hoping for something like what you answered. And Put is indeed better then Export[..., "Package"] and breaks less things. It still breaks a few though ...
Maybe my question is just misguided and I should set up my "build environment" so the palette is always generated with a v8 kernel.
Then it will be more clear that 7 is not supported for anyone who tries to install it on 7. (I believe it works on 7, but I can't test)
 
Better than accepting defeat and using regexp to postprocess the saved .nb
maybe it's kind of easy though.... just change "$Version" into "8.0 for Linux x86 (64-bit) (October 10, 2011)"
or whatever
 
@ssch Are you saying that it writes the value of the kernel variable $Version in the .nb and we can change that temporarily with Block? I thought this was hard coded in the FE
 
@Szabolcs nono, uglier than that, save, read in as pure text, StringReplace, save again
but if Block can override such things surely that'd be better
 
5:45 PM
aw, $Version is locked, I don't want to have to go around that
 
rawnb = Import["/tmp/test.nb", {"Text", "String"}];
Export["/tmp/test_nfev.nb",
  StringReplace[
   rawnb, $Version ->
    "1996 \[LongDash] Microsoft FrontPage 97 (version 2)"],
  {"Text", "String"}];
sadly complains that it was made with a more recent version :(
 
hmm? Strange. Must be because of the comments at the beginning.
 
probably because of 1996 > 8
 
Please do undelete it though
 
I put it in the first answer
 
5:52 PM
Okay.
 
didn't want to crowd up with half-working stuff :)
 
@RolfMertig the need for player pro is exactly the issue that I feel they messed up on. What I want is an equivalent for a jre for my mathematica code ... sadly without player pro and its impossible licensing terms I just don't distribute my code to anyone without mma.
 
6:15 PM
@Gabriel, are you aware that you can now get the Enterprise version of mathematica and sign CDF documents that have all the features of player pro (import, export, etc) but run in the free player?
 
7:08 PM
@David I have heard this mentioned but it is difficult to get information about how you do this. I would really love this if it is not prohibitively expensive.
 
@Gabriel It is shown in their store now for $7000. The blurb on the store says "CDF files deployed with this option offer enhanced capabilities, such as data import and export, use of arbitrary input fields, and file saving."
 
7:37 PM
@David cool. Okay, so maybe I am not so angry at cdf then ;-)
 
acl
8:09 PM
funny how WRI is simultaneously accused of catering only to academia despite them charging roughly what a single researcher would budget for all non-personnel expenses for a year (or even more in most cases) in order to allow them to run things outside their own environment
 
The other someone was discussing CDF and what functions can be used and not used. I'd seen a list but couldn't remember where. Just came across it when about to answer another question. The list of "dangerous symbols" is contained in MiscExpressions.tr which can be found in systemResources>>FrontEnd>>TextResources
@Gabriel the way it works is that you buy a CDF Enterprise. This "pro-signs" the CDFs you make and this signing unlocks features in the free CDF player -- meaning that someone with the free player can now import/export etc.
The additional catch I only found out about earlier this week is that in addition to the cost of CDF enterprise you have to pay Wolfram royalties on the apps you make and sell.
You will notice that import is listed as a dangerous symbol.
So maybe the pro-signing replaces this list with an alternative list???
 
8:41 PM
Seems the suggestions bar and all that is actually quite tweakable
 
Does anyone here use SystemModeler? I'm just curious. The other day I got an email about it, but I don't think it would be that useful for the work of most people who frquent this chatroom.
@Rojo How?
 
I'm foreseeing @Szabolcs uploader button after graphics
 
Good idea :-)
 
@Szabolcs It will take some free time or help to understand the code, but it is quite readable and organized, and even most of the functions have a usage message
in the Predictions and predictiveInterface context
@Szabolcs Try this for example
MathLink`CallFrontEnd[
FrontEnd`AttachCell[First@Cells[], Cell["Febo asoma"]]]
 
Have you seen this? It helps a lot with spelunking.
2
A: What is the most convenient way to read definitions of in-memory symbols when we don't have the source files?

Simon WoodsI have been using this. It's mostly Leonid's code from the stackoverflow question you linked to, but it uses FullDefinition instead of DownValues. Symbol names are printed without any context, but the full symbol name is put into a Tooltip so you can always find out what context a symbol is in. ...

 
8:45 PM
No I haven't, I used a couple of my own ones. Thanks
Quite useful stuff
 
I see the whole thing is an application (browseable source in the Applications folder). Some of the files are .mx though.
It seems all the good stuff is in the .mx files.
 
Well, they get loaded, so no problem
 
Which ones have usage messages?
 
I'm happy to see that it's all based on FrontEnd`AttachCell
which we can use for whatever
 
@Szabolcs I have SystemModeler but have yet to use it. One of the things to do over the Xmas break is to play around with it. I have had it shown to me with some basic demos and it looks good for users you may be working with e.g. large systems of coupled/nonlinear DEs
 
8:50 PM
passing it closing events...
 
the two way integration with Mma is nice.
 
@Szabolcs do ?Predictions`*
and start clicking
 
@Rojo The only thing there is Predictions`MakePredictions
oh ... I see now. It wasn't loaded yet.
 
There you go
 
@Rojo Have you already figured out which function decides about which predictions to show?
 
8:55 PM
@Szabolcs Not clearly enough, but I think it's Predictions`PredictionRule
Careful, its big
Based on the input and output pair of cells, and some "type".
Ever saw that "bla is treted as a vector, treat as a permutation instead"? That would be the type I'm guessing
Haven't yet looked too much
Predictions`Private`SemanticTypeNames[] seems to have a list of type names
 
now the only question is: what is name? the thing that looks like {_,_,_,_, ... True ...,_,_,_}
 
@Szabolcs According to the usage message, it is just an identifier to avoid overriding rules that look the same
@Szabolcs I didn't quite understood that code either, that didn't look like the usage message described, so I moved on
Too many things to look into
 
@Rojo Do you want to post a question about extending the prediction interface and post some basic findings as an answer? Just need to get people interested a little bit
 
Cuter than a panel ExperimentalFramedLabeled[Plot[Sin[x], {x, 0, 10}], "Febo asoma"]`
@Szabolcs I have to leave in 5, feel free to do it yoruself. Otherwise, perhaps in 3-4hs
 
9:24 PM
1
Q: How can we extend the predictive interface so it can suggest new operations?

SzabolcsIntroduction Mathematica 9 has introduced a new predictive interface. For each output, it suggest possible actions the user might want to take. How can this interface be extended so that it will suggest additional, possibly user-defined, actions? The starting point @Rojo mentioned in cha...

For those who are interested. Of course these are undocumented for a reason and whatever solution we can come up with might easily break in future versions. Also, in the experts live session it was mentioned that making this extensible is something to consider, but it's impossible to know when (if ever) that will happen ...
 
@Szabolcs where did your question go?
 
@MikeHoneychurch I just deleted it because Leonid pointed me to this one But I am changing my mind. For that question, I would implement a separate suggestion bar from scratch while I'd keep my question for customizing the built-in predictive interface.
What do you think?
 
9:41 PM
I never got to read your question but i guess there was a lot of overlap??
 
acl
@MikeHoneychurch how so? I'd have thought that in that case it makes sense to construct them programmatically (which is what I do). can I see these demonstrations somewhere?
 
@acl as I indicated i was guessing cause I never read @Szabolcs post before it got removed ...I'm assuming you are referring to my comment immediately above??
 
acl
@MikeHoneychurch No I was asking about an earlier comment to the effect that system modeler would be useful for people working with large sets of DEs
 
@acl oh ok. You can create a model of your system with drag and drop widgets and when complete and checked for validity you may well have created the equivalent of hundreds of coupled ODEs (for example). So while you could obviously write down all the equations for a particular system in Mma this makes it all a lot easier.
And when linking from Mma you can change model parameters from Mma.
 
Hello @jrhodin. Funny that you appear just half an hour after SystemModeler got mentioned ;-)
 
acl
9:56 PM
@MikeHoneychurch right I see, I guess we have different kinds of systems in mind
 
@jrhodin it looks like you r a modelica guy? Maybe you can respond to @acl??
@acl what did you have in mind?
 
@Mike Do you do Mathematica consulting? If yes, you must have met lots of people who use Mma. Have you met any physicists who use SystemModeler or something similar to that? I am just curious.
 
acl
@MikeHoneychurch huge systems of coupled nonlinear ODEs, which however are very simple to write down (you were probably thinking of large systems of very different ODEs)
 
@Szabolcs Yes. I did a roadshow with a guy from Wolfram a few weeks ago who was demonstrating it but I haven't had time to get familiar with it yet. It is something i want to try and do over Xmas -- take some baby steps anyway
@acl best have a look at some of the examples. Also modelica.org
@Szabolcs second part of your question: in academia yes. outside academia no.
@Szabolcs third part of your question -- because I do not know SystemModeler and it is relatively new I have not sought out people who use it -- though I am aware that people use these tools (Dymola, MapleSim, Simscape, Simulink, Vissim)
I did a project for a guy last year in which modelica might have been useful but never got any assistance at that stage and I do not know modelica. So the guy has built something from scratch with C++ debtdeflation.com/blogs/2012/12/16/help-kickstart-minsky
 
@MikeHoneychurch Do you know how one would go about becoming a WRI certified mma instructor (some familiar names and faces there...)? Do those folks get paid for teaching mma (wherever they are)?
 
10:10 PM
got to run right now or we miss the bus and my wife will kill me!
 
Contact WRI and ask them to certify you (maybe the contact link on the page you have linked). They will ask you to teach a couple of units of their course in front of (or skype) an instructor. If you want to get paid you would either have to hang out a shingle and run your own courses or work with someone who runs training courses.
 
@MikeHoneychurch But if a company does not sell the app?
 
@MikeHoneychurch Thanks.
 
@Rolf not sure what you mean/asking?
 
CDF pro
 
10:17 PM
Sorry still unclear. What specifically about CDF pro??
 
@MikeHoneychurch You said one has to pay royalties... Royalties typically implies a cut of the profits, so Rolf wants to know what the deal is if the company isn't selling the app.
 
So say, company XX has M-Enterprise, signs CDFs as PlayerPro, but they run in Player. Does WRI want royalties on that?
 
ok. sorry. I would assume if you are giving CDFs away for free then you have no royalty to pay. I haven't seen a license agreement and I am guessing that their model is still rather fluid but my understanding is that royalties would be on apps that are sold.
 
My point is: the license is not clear on their web-site. They say one should talk to them. Well, but what if I do not want to talk because it is not clear and I just want to know the legal implications. It is funny how complicated this whole CDF things has been, is , and seemingly continues to be ...
O.k., App-sell is fine, like Apple-apps. But the point is: what if it is commercial, but not sold in the wild? Uh well. Not that important.
 
One can only claim a royalty on revenues earned. If no revenues are earned (you give your stuff away) then by definition x% of zero is zero.
...but as above I have yet to see the license agreement and would not be surprised if it is not even fully drafted yet.
 
10:24 PM
Do you know when WRI will release a M9-Player?
 
I have been told but I am not sure if it is public info ...sorry. ...but it would be reasonable to assume it would not be long after 9 ...right
FWIW I think WRI should not be charging royalties on retail sales but rather on wholesale sales, or better still set up an app store and take the retail margin. Just my 2 cents.
 
WRI should understand that they will never ever get substantially more business if everybody has to talk to them first for legal things. So yes, the easier the better. App-store would be fine with me.
 
It might be good if we could get someone from Wolfram to be on here one day for a Q&A chat.
 
Call Andre ...
 
Andre was who i had in mind :)
 
10:32 PM
Yes. good idea.
 
But i would guess they would want to have all this stuff released before doing a Q&A. I'll email him and aks him about doing it when it is all out there.
 
I am running out of space on my CDF demo UI to add more useful information to display :( reached image size limit. Have to be really imaginative now to figure how to add more things in such small space
 
But then it is too late to influence him ...
 
If you want to influence him then you should email him. that is what I do :) (or offer feedback etc.) I'm guessing you know Andre. He is very helpful and responds fast
Andre just emailed me and said there will be webcast Q&As.
 
10:48 PM
@RolfMertig totally agree with the fact that if the cdf rules are not clear, if it is necessary to contact just to understand a little more than the basics, a lot of business gets loss. Most projects start with small ideas. With someone thinking on the possibilities. If things are not clear, imagination gets lost...
I've been complaining on this model (and other details) from some time...
 
@P.Fonseca Let's hope Andre K. will listen. He is working for ages at WRI. And he is cooperative.
It would probably already be nicer if WRI allowed downloads of the CDF Player in a similar way as Adobe does: no email needed. Just download, install, run, enjoy.
 
Today, during a management training, I started thinking on how to improve my automatic reporting document (some automatic graphics and statistics that are produced directly from the info coming out of the erp). Then, going a little further, I thought on WRI idea of repports (always thought of then as out of reality... but I gave some minutes of thought). My thought ended up with: why would I ever pay 9000€ to be able to distribute reports that access external source of information?
I can add the information on a free CDF, but then it becomes a confidential security problem...
So, is there realy such an advantage on reporting on cdf that deserves paying 9000€ when compaired to a simple pdf?
 
0
A: Evaluating a Recursive Expression

sschA simpler example where I tell Mathematica to print what it's doing. I ClearAll[n] to make sure n has no value set. ClearAll[n]; f[0] = 0 f[n_] := f[n - 1] Block[{$RecursionLimit = 20, $IterationLimit = 20}, TracePrint[f[n]] ] See that there is a bunch of f[-10+n] etc, but since n is not ass...

I can't figure out how to escape the links to $RecursionLimit and $IterationLimit
 
@P.Fonseca, I do not know about 9000 pounds and such. By with CDF users can interact with the report. Here is an example of what is called automated reports using CDF wolfram.com/cdf/uses-examples/automated-reports.html
 
My conclusion is no. And if I need to pay 9000€ to get a report locked on proprietary technology, then I prefer to the the Html way...
The question of the 9000 concerns the pro advantages (mainly, the access to import export, etc)
 
11:01 PM
@P.Fonseca if you were distributing a report wouldn't you have the data stored in the CDF and therefore distribute it as a free CDF?
 
@P.Fonseca, can you do all these things that CDF does with HTML? if so, then yes, use HTML ofcourse :)
 
If i store everything that is analysed inside the cdf, i'm distributing raw pure information on an easily transmitable file...
 
@P.Fonseca it will also be accessible in a pro-signed CDF so if security is an issue then CDF is not for you -- no matter what type of CDF.
 
@MikeHoneychurch the pro has all capabilities of access to external databases, so not need to include information on the file
@NasserM.Abbasi can't we do everything on free technologies? If we can do the entire office suit on internet, we can do most of intface and calculations on html (backed by other languages...). So, it is mot a question of being able to do it, it is a question on the heigh at which WRI has placed the bar to make things usable.
 
@P.Fonseca sorry but I am finding the discussion to be moving around a lot. Step by step what exactly is the problem with CDF enterprise?
 
11:08 PM
(meanwhile, sorry for my eventual typo mistakes, but I'm having problems with tha tablet...)
 
Above you have compared 9000 euros CDF to a simple PDF. The PDF will have some data displayed. If there is no security issue with that data in the PDF why can't that data be in a free CDF? Sorry, just not seeing what the issue is?
 
The price is huge to be able to do the most simple stuff. I know that I gain access to do things that are much more advanced than simple stuff, and even distribute serious apps, but since most interesting ideas start simple, I'm not seeing how simple can start at such a high investment. I know that I can try things out before investing, but why would I even try to think on going that way if I my be going to a dead end?
Obviously, if i do it in pdf, i do it static. That is, procesed information. That is, you get what you see, and not what you would like to see... The advantage of the cdf as a report is that you can change the process of the information, but for that you need access to the raw info. But you will probably don't want it to get out of hands
 
I think we need to partition apples with apples and oranges with oranges. If you would ever consider distributing a free PDF then I do not see why you cannot use a free CDF. For Enterprise CDF then any potential user has to weigh cost vs benefits. It won't be for everyone.
 
I know that it is not for everyone. But my opinion (and it is exclusively my opinion that I am stating here), is that the bar is too high for it to take momentum
 
...and if a free open source CDF pro existed wouldn't your data security issue still exist? I mean I could access all your data via my open source CDF interface. I just think we are going around in circles.
 
11:17 PM
No because i can keep it inside network access
 
So why not use webMathematica then
 
Even more expensive ... But yes, webMathematica is the way to go for really secure and extendible distribution of easy-to-use applications.
 
First, it is not as intereesting as the cdf app. Then, and again, most big ideas start simple... and both enterprise edition and webmathemtica have too much mass to put in motion. There's no let me try this ans that, and loose 5 minutes today and ten tomorrow to see how it goes... It instantly becomes, lets contact WRI, explain what is the purpose of the project (that is going into brainstorm...), and why not install a server... For a technology that is perfect for prototyping..
 
BTW: What is wrong with using Encode even in the free Player? Is this forbidden? It does work.
 
Correct me if i'me wrong, but if i need to decode it to process inside the own cdf, it is also available fro everyone else.
 
11:27 PM
You can test out webMathematica. I have done consulting projects involving webMathematica with a reseller here in Europe and the resellers let customers try things out first.
@P.Fonseca: How would you do that? The CDF is closed, evaluation wise speaking.
 
@RolfMertig I haven't seen recent pricing but though webMma was considerable cheaper than enterprise CDF??
 
@MikeHoneychurch No. You have to buy webM for three years at least. I promised the reseller (candiensten) though never ever to talk about money, license costs, etc. ... And it may very well differ from country to country.
 
And again, lets contact this and that, and discuss experimental scenarios. It is to much mass just to imagine things, and do brainatorming. I'm talking of trying ideas, and not in preaproved projects with starting budgets.
It all starts samll...
 
Now, the real fun is when you generate dynamic CDF's from within webMathematica ...
 
Small...
 
11:32 PM
@P.Fonseca Well, difficult web-sites are not small projects.
 
But ideas start small. If things match expectations, we invest. If we atart by testing on alternativa technologies, we will go away from WRI tecnology
I would love to know how many enterprise or player pro editions were sold for corporates.
During the last 4? years...
If it is a consideravle amount, then i'm completly out. If it is a small quantity, then, probably, it will still be a small quantity within 4 years... And the technology doesn't gain the needed momentum.
 
If I understand things right PlayerPro and FreePlayer will merge in Version 9 into just one "Wolfram Player" or "CDF Player" or how they going to call it. So there will be no PlayerPro sales anymore soon. But I predict that the Enterprise sales will go up once people understand how cool "Enterprise enhanced" signing is. I think WRI should give all Stackexchange members with more than 1k rep or so free upgrade to Enterprise ...
 
Since strategy has been changing, I'm gessing that it was a small mount, and since i'm on the corporate world, one of the reasons may probably be the one that blocks me. But i admit that i may be completely wrong, which is why I'm simply staring my opinion.
@RolfMertig thank you for having included my small rep category on the promotion :)
 
Who knows. I am glad WRI moves in the right direction. With respect to prices: that is always hard. If something is too cheap, your boss thinks it is not worthwhile. If it is too expensive, it is prohibitive. Contact your Mathematica reseller (or the UK office) and tell them you want to test things out and you have help from us here on SE. I am sure they will be nice to you.
@P.Fonseca I don't have that much rep either. Sometimes I don't have time. Sometimes I am too slow compared to the superstars here ... Probably I am getting old.
 
I'm always too slow
I'm clearly old compared to average here...
 
11:48 PM
Most questions are asked and answered by the time I log on in the morning.
 
You can see by my eyes on my sympsonized photo...
It is late here. See you...
 
@P.Fonseca good night!
 

« first day (335 days earlier)      last day (4155 days later) »