Altair RISE

A program to recognize and reward our most engaged community members

Nominate Yourself Now!

"Version 5 documentation

Hi RM,
I'm new to RM but not to the opensource community.
I must say I have never encountered an opensource project with such a steep learning curve and so little documentation - close to none.
Since there are no examples anywhere to be found which do not contain deprecated methods, I have to figure it all out on my own.
Meanwhile the API is no help either.
For example, in order to not load the jdbs, the API (and Wiki) suggests setting some properties to false.
However this does not help and looking inside the code (with a debugger) I see that reading these properties has been commented out!!!
So please understand my frustration and helplessness.
It would be a TREMENDOUS help if you could please post a full working code example that does something, for example the code for one of the examples in the tutorial (but Java code, not reading an xml file), together with the data file.
The more of these examples you post, the better.
Please help your faithful users!!
Thanks
Carmit

Find more posts tagged with

AI Studio

Text Mining + NLP

Accepted answers

All comments

haddock

Hi there,

Because you are new to RapidMiner you may not know where all the documentation is, for instance..

http://rapid-i.com/component/option,com_virtuemart/page,shop.browse/category_id,5/Itemid,180/

I am an absolute moron, and yet have managed to write my own operators, so for a smart fellow like yourself it should present no problems at all.

Have fun!

carmit

Thank you for your kind reply, leading me to conference proceedings and a 43-page long manual which is simply NOT ENOUGH as a start for a total newbie.
I have spent close to a week trying to figure out how to solve this mystery which is RM, and I can't afford to spend any more.
Opensource projects of much lower complexity than RM have plenty of up-to-date documentation & examples.
Perhaps coming from the well documented version 4.6 it's not very easy for you to realize how impenetrable RM is.
So, can you please post a few examples of running code, complete with the input files and with no deprecated methods?
What would take you 5 minutes to post would take hundreds of potential users days upon days to figure out - and only by painstaking trial and error or reading the RM source code (since no other available resource is currently up to date).
Thanks
(And please try to not to grade my intelligence)
Carmit

haddock

Carmit,

Mmm. Did you buy and read the manual I suggested, or DO YOU JUST LIKE SHOUTING? If that is beyond you then you could try elsewhere, but you must bear in mind that others have managed on what I have pointed out to you.

Good bye.

IngoRM

Hi,

I must say that I totally agree on what Haddock has said. Beside our multi-hundreds-of-thousands-dollars-license-costs-software-for-free we also offer

An English and a German Manual of about 150 pages for which we got a massive amount of positive comments - again for free
More than 50 videos from us and many more from community members - again for free
A developer manual for extending RapidMiner for the small amount of 40 bucks
A second (planned) one which will be released shortly (look here in the forum for more details)
This forum here with many helpful people willing to help you - if they know your problem.
A Wiki containing a description of all operators - and which could be improved - also by you!
Hey, and finally it's open source - of course you also have access to the source code and the API

Not enough? Well, time is money. Get in contact with Rapid-I, ask for an offer for our Support Subscriptions and I am sure you will never have to spend a week searching for a RM-solution again. Please forgive me that I don't get in panic about the one missing piece from the 4.6 documentation - which is, as I have stated above, already planned anyway.

Totally lacking? Well, if you say...

Cheers,
Ingo

haddock

What Ho Ingers!

There is an angle to this which should be made more explicit, which is about why folks should pay for help on Java coding. Most extensions are created by people with a particular need not already covered by RapidMiner. The license arrangements require that, unless you pay, you will have to make your published extension source as open as RapidMiner's source.

As a consequence very few extensions are published, except by RapidMiner, and help in making them can only really benefit the maker. So if makers want help quick, they will have to pay, as with any personalised service, or hope to get lucky by annoying the passing traffic.

This may not be so bad, as extensions also need upgrading - what happens when the publisher of an open source extension decides not to support or upgrade? Whatever else ( and there is much ), Ingo and his band of merry men demonstrate amazing persistence in supporting the extensions, and that is important if your project is.

carmit

I'm sorry if I sounded ungrateful

You ARE doing a great job!!
Carmit

IngoRM

You ARE doing a great job!!

Thanks :-*

Well, I must admit that the "totally" in the headline indeed has sounded a bit ungrateful

@Haddock:

I am not 100% sure if you are argueing in favor or against such whitepapers which can be buyed for a pretty small amount of money. Just to make sure, here is my argument why I think it's fair and even a good idea

why folks should pay for help on Java coding.

Well, for the same reason you probably also have to pay for a Java book. Or a Java training course. Or... Most developers, especially in commercial settings, are quite used to the fact that they have to buy additional information if they want to code for a specific solution. And this is completely independent of open or closed source solutions.

The reason why we have to charge for this is quite simple: documentation about development issues takes by far the most time and unfortunately it has to be updated even for the slightest changes. At the same time, this large amount of work has to be done for only a small group of people (let's face it: there might be hundreds of thousands of regular RapidMiner users but probably only like a couple of hundred RapidMiner developers). Given the amount of third-party extensions for RM < 5, I highly doubt that the number would be much higher if the whitepaper was for free

And it might also be a question of commitment: When you consider the price, you may also want to consider what you gain by using the software, by receiving the whitepaper, and by being able to extend RapidMiner according to your wishes. And maybe you also start to think about what you might be able to contribute...

Is 40 Euros really that much in comparison to what you gain?

Haddock, I know that you don't think so but I hope that these thoughts might also help others to better understand this decision.

All the best,
Ingo

haddock

Ingo!

Twice in one day - we can't go on meeting like this!

Just to be clear...

You should charge as much as you can get for papers that support Java coders, because only they benefit.

You will starve by servicing folk who cannot pay 40 Euros, but can afford a computer and Internet access.

There should be a mandatory splash screen, pointing to the FMs ( Fantastic Manuals as Steffen points out ).

Perhaps also a Forum Policy which we can link to as a stylised form of RTFM reply.

Richy

Hi,

Since there are no examples anywhere to be found which do not contain deprecated methods, I have to figure it all out on my own.

As you can see on Ingo's signature :

We encourage everybody to share RM workflows on MyExperiment.org portal. Install the RM Community Extension via the RM updater and follow the simple HowTo on http://www.e-lico.eu/?q=node/226

You can find many examples on MyExperiment.

Regards,
Richard