Forum TI-Planet.org

Dans un sujet sur le développement et la programmation, notre cher Google-Group Tinspire nous accuse des pires maux du système éducatif américain!


On s'est fait traiter au départ de vulgaires programmeurs de jeux, à cause desquels les élèves n'écoutent pas et n'apprennent rien en classe...

Et quand quelques exemples de programmes éducatifs ont été donnés, on s'est fait reprocher:

• de faire des trucs inutiles car l'OS fait déjà tout
• de faire des programmes non fiables qui plantent ou donnent des réponses fausses

En conclusion, Wayne dit qu'il est donc parfaitement normal que TI limite les possibilités de programmation/développement au minimum, afin de nous empêcher de faire des bêtises...

Wayne wrote:The major problems with third-party mathematical or education programs
are the lack of accuracy, trustworthiness and support. There are many
thrid-party mathematical programs for the TI89 and TI92 hardware that
are published on ticalc.org and other websites by well-known
developers in the third party community. Many of these programs are
not needed because they duplicate built-in OS capabililty, are
inaccurate and vulnerable to random failures, or are mathematically
wrong in their entirety. An experienced mathematician can test these
products and determine which should be trusted and which should be
discarded. Unfortunately, the majority of the people who download
these third party programs are not equipped to make that
determination. As a result, inexperienced users download these
programs and think they are getting mathematically sound answers from
them.

If TI retains control of the development of applications, we can
expect at least some level of trustworthiness and support from TI. If
third party developers of mathematical or educational programs commit
to that same level of accuracy and trustworthiness of their creations
or to support their programs if they are found to be otherwise, then I
say let them have at it. If they not willing to support their
creations properly, then control of the platform should remain with
TI. In my experience, most third-part developers do not commit to the
accuracy and trustworthiness of their programs or to their support if
found to be mathematically wrong in some manner.

http://groups.google.com.au/group/tinsp ... hl%3Den-GB

Je trouve la toute dernière accusation de Wayne particulièrement injuste !
Quel développeur digne de ce nom se casserait le cul à programmer quelque chose et ne prendrait même pas le temps de tenter de corriger si on lui montre que son "produit" ne fonctionne pas comme il devrait ?
En général, c'est plutôt l'inverse qui se produit : lorsqu'un programme est plutôt conséquent et compliqué, il est difficile de trouver des testeurs suffisamment aguerris pour découvrir s'il y a une faille ou non... alors que l'on aimerait avoir une autre opinion que la sienne.

Par ailleurs son point de vue est extrêmement prétentieux et même fallacieux : ça se saurait si TI était fiable à 100%...

Le temps qu'on m'accepte sur le GG, voilà ma réponse que je garde en mémoire :

____________________________________________________________________________

Hi Wayne.
I agree with you on many points but I think you're going too far in your description of the TI-Community.
Fisrt of all, TI gives us capability to make our own programs thus we make our own programs.
If TI doesn't give us functionalities, it is a challenge to the community to make them.
If there is not a Computer Algebra System on a TI-nSpire, it is for some reasons, and "why" some users like Critor or Bwang make engines like CAS for this TI-nSpire, it is to serve the community.
Furthermore, we live in a Open midding community, where everybody gives their own knowledge and experiencies to help other people.
If an "inexperimented user downloads these programs and thinks he/she is getting mathematically sound answers from them", I'll answer you that this user will not able to solve the mathematical problem by his own and it is his fault, not the community fault. Thus you could not accuse the community to create free stuff because they're are not reliable. It is to the user to make the choice.
Finally, all inexperimented user are looking for "magic programs" and you know better than me that those programs don't exist.
When TI gives us all the things that this user need, they will never appear those kind of program that you describe here :

Wayne said _________________________________________________
inaccurate and vulnerable to random failures, or are mathematically
wrong in their entirety. An experienced mathematician can test these
products and determine which should be trusted and which should be
discarded.
___________________________________________________________

On the other hand, I can't bear that TI wants to control the user mouvement.

Think that the Nspire hadn't third party ASM
Think that the TI-community solved the problem with Ndless
Think about the power of the ASM, think that with this we could make any application we want either TI never gives us a SDK on Nspire
Think that the OS 2.1 refuses downgrade.
Think that the TI-community solved the problem with Nleach.
Think that we could not make 3D on the build-in OS
Think that the TI-Community solved the problem again, and not by an "unexperimented developpers", but by a mathematics teacher.
I have today many examples because of the TI-Community which never stopped in front of the TI-challenges.

We assume the responsability of our calculator or not. We assume what we download or not. Never a developper will say "THIS IS THE BEST PROGRAM EVER" if it is not.
Ndless is unfortunately awesome, and the user HAVE to assume what he does. It is the same thing for a computer. You assume what you install on your computer.

Good regards,
Levak

____________________________________________________________________________

Levak: tu ne pourras discuter avec Wayne qu'en privé, vu qu'un soi-disant modérateur, qui a lui-même posté des conneries dans cette discussion, a fermé la discussion.

Je discute en privé avec le Wayne. A le lire, MathTools contient plein de bugs - mais le seul qu'il m'a listé, pour le moment, n'existe peut-être tout simplement pas: la doc et le programme donnent un résultat différent.

Ceux qui ont accès à  Maple, Mathematica ou d'autres CAS similaires: quelle est la valeur de Zeta(1-i) ?

Selon Wolfram Alpha ....

http://www.wolframalpha.com/input/?i=Zeta%281-i%29


0.582158059752003648199463167914259201877989316826534645721... +
0.926848564330807076536424313917500774053454893873943427684... i

Mouarf, je ne pensais plus à  Wolfram Alpha :

Donc le "bug" qu'il m'a indiqué n'existe juste pas.

Et j'ai édité mon précédent message :;):

Oui, et peut-être de se faire bien voir obséquieusement par TI .....

Heu, en gros seul TI crée des programmes fiables et sans failles ? haha, la bonne blague...

EDIT : bon c'est p'tet un peu exagéré mais c'est comme ça que je le comprends, qu'il faut laisser le contrôle total à TI pour éviter d'avoir des programmes pas fiables et avec des bugs. C'est vrai que certains programmes sont peu fiables mais de là généraliser faut pas abuser non plus...

TI fait des bugs, aussi: http://www.ibiblio.org/technicalc/buglist/bugs.pdf
(je sais que tu le sais - je le poste dans ce topic et un autre pour augmenter le page rank de ce PDF : )

et pourquio pas le poster sur le google group aussi ?