Is it the end for Cognos Planning . Is Future Cognos TM1

[cogplan]

Hi Techies
              I come from engineering background and I am working on Cognos Planning.As as far as market and jobs goes Planning clients are switching to TM1.Is it the end for Cognos Planning. and do I need to move on to TM1 and learn it asap. Can you please suggest me how to improve my knowledge in this space both functionally and technically . so that I can build a solid foundation and move on with my career and job prospects..

            I am sorry for posting this in a technical discussion forum. Please advice

Regards
CogPlan

[SomeClown]

Short answer, yes.  IBM will say otherwise, but look at where money and time are focused.  It's only a matter of time.  It would seem IBM is not admitting this because they don't want to lose Planning clients to competitors.

[ericlfg]

There will always be speculation regarding this topic until the Product Managers and Executive levels release something stating the -Actual- path.  I can't imagine, based on how many companies and organizations use the product, that IBM will one day throw out the Planning product.  At least not in the immediate future.

If there is still revenue being generated that offsets the costs of development, they'll continue with it.  IBM, like any other company out there, is looking to continue making money -- no matter how diversified their product lines are.

[cogplan]

Thanks for the reply Techies. :) :) :)
        The confusion sets up for a techie like me, who's in the midst of learning new tools specific to EPM. As far as this discussion goes , I sense the need for me to learn TM1 and start working on it.
Certainly company wise be it IBM,Oracle,SAP all are here to make money.
         There's always a talk about intregrating Contributor with TM1 and keeping certain functionalities of Cognos EP. Can you please throw some light on high level differences and the complexity in TM1. As in most of the forums they discuss about the complexity of TM1 as to EP's Simplicity.

       Hence I would like to know , can we download TM1 developer edition from IBM and practice on that.

Regards
Cogplan

[cogplan]

I need to add something to this.
       Why is there Cognos Ep 10 version ..

[SomeClown]

Re: Cognos 10

Compliance with infrastructure (e.g. Windows 7)
Branding (almost everything is 10 - even DB2)
Hot-fix/patch rollup

Yes, barring statements from IBM, it's all conjecture, but neglect and mismanagement are just as effective in killing a product  (look at the sales numbers and sales organization if you want to see where real organization investment is made)

How do you like all the new features in C10 Planning?  http://www-01.ibm.com/common/ssi/cgi-bin/ssialias?infotype=AN&subtype=CA&htmlfid=897/ENUS210-349&appname=USN  Lots of things there that were not in 8.3/8.4

[ericlfg]

Can you ever really learn -enough- in IT?  :)  Just as companies have to stay competitive and release new and updated software (whether that is within their primary brands, or through acquisitions to reduce the number of competitors (Cognos purchasing TM1 as an example)), we must also learn and update our skill sets to remain marketable.  Long way of saying you should learn both Planning and TM1, to at least a basic functionality and understand the raw differences..   

Can you please throw some light on high level differences and the complexity in TM1. As in most of the forums they discuss about the complexity of TM1 as to EP's Simplicity.

Most of the complexities of TM1 come from the "Rules" areas.  These are hand written and you must have a very good grasp of what you're doing in order to make this area work.  Rules are loosely the same as d-links in Analyst on the Planning side.  In TM1 where you have to write rules manually, in Planning you have a GUI with a good representation of how the data will flow.

Why is there Cognos Ep 10 version ..

SomeClown has covered this well.  I'll just speak towards some personal experience with it, in that I have seen some substantial improvements to reconcile performance as publishing performance.

Cheers,
Eric

[SomeClown]

I agree that C10 is a great release in its improvements.  So much so that I'd recommend that anyone using Contributor should seriously consider upgrading, even from 8.4.1.    It's unfortunate and very disappointing that the inappropriate positioning and marketing for TM1 is weakening the market position of both products (IMO).

[cogplan]

Thanks Some Clown and Eric
                Thanks for yur suggestions and also for your insights on both the products.
Business or money making strategy has  taken IT this far.. Yeah I agree the point which Eric made . You have to update yourself to make yourself marketable..
               I have not got a chance to look into Cognos 10 suite. esp Planning..
                 Its been a nice discussion thanks everyone .

                Bottom line is IBM is in full force with TM1 and also using planning as a crowd puller to TM1..

Regards
Cogplan

[Rutulian]

My 2 penn'orth:

TM1
Very powerful on one box
Limited scalability
Pig to use
Consultants who can write rules even rarer than Cognos EP Modellers
Not so pretty on the UI side, not really SOA
Can have a 'true' web-only client (I believe, corrections invited) with no local install needed

Cognos EP
Distributes calculations (when Contributor is involved)
UI that effectively gives you click-by-click ETL into OLAP cubes
Just about shoehorned into C8's SOA architecture by 8.3 & later, needs to ditch Analyst for something Web-based
Client installs without local admin permissions

I continue to be surprised that TM/1 is pushed, as it seems to me the best technical solution to all of this would be to slap TM/1's calc engine behind Contributor's UI.  Not just the mungey Contrib-client-for-TM/1, but sticking that calculation engine into the Job Servers and having something that turns Contributor XML into TM/1 models.  This would probably take a few years to develop.  IBM/Cognos has had Applix for a few years now.  I'm sure the internal politics of killing 2 products to make 1 better than either (not to mention migration paths for existing customers) are unfun to navigate, though.

What interests me from a far-too-technical-to-be-useful perspective is whether or not such a product would also use the TM/1 engine in the client, or if J would live on there - I think TM/1 is C++ based, whereas EP's history is in functional languages like APL and J... if you were into intellectual self-abuse you might try to implement a functional domain language on top of TM/1's engine.  It'd be a big roll of the dice, but the ability to iterate in future would be much quicker than a one-shot EP to TM/1 conversion and maybe we'd get the ability to write custom BiFs again!

Interesting chat, thanks all - I'd be especially like to hear if anyone knows of much the TM/1 UI can do that couldn't be fairly easily added into something like Analyst, but web-based.

Alexis

[AWILDE]

Hope you find my response useful:

IBM have now put TM1 ontop of contributor in v9.5.

Pig to use - I am 3 months in to using planning and have used tm1 for 4 years but would say Tm1 is equal on how to use, but more flexible in designing.

Consultants to write rules rare - There are a number of consulting companies and you can normally get a consultant out within a week from experience.

Not so pretty?? If you use Executive Viewer or TM1 web the reports look amazing.

What TM1 does can not be added to analyst as it users Feeders to control what cells are calculated this is known as Stargate Algorithm.  It also users hireachy in dimensions (dlists in analyst).  Rules are written by cube area not within Dlists.  It is very different to analyst.

[Rutulian]

Thanks AWilde,

I should have disclaimered at the beginning by stating this is spouting off the top of my head, not a considered and balanced opinion.  I haven't had much hands-on with TM/1 (only installs really, where it's very friendly indeed).  'Pig to use' was certainly a poor choice of words - just an impression I've picked up, but that's largely from EP consultants hard-wired to think dimensionally.

TM/1 is definitely more flexible with what can be done with rules and individual cell manipulations.  More power, less ease of use for nontechnical users.

We've not got true Contributor-on-TM1 yet though - what we have is the Contributor Web Grid handling data collection for TM/1.  What's really exciting (from an EP-biased perspective) is the possibilities to have full Contributor chopping up the workload along the e.list dimension, spawning multiple TM/1 instances on job servers to crunch the numbers.  That or TM/1 behind a web-based Analyst-style builder to overcome the limitations on the calc side.

Maybe we'll get a one-shot tool that can parse Contributor XML models into TM/1, intelligently placed SkipChecks and all, and then still allow Rules to be added at that point - sounds like a maintenance headache but would take the pain out of multicube model builds with the 'noddy ETL' options of D-Lists and still let you get into the guts of TM/1.

Both products also deserve credit for D/T/P migration options - TM/1 seems very easy to move instances between servers by copying the model folders, while EP's deployment functionality is a bit more hassle it lets you move partial models and do things like remap user populations which can save a massive amount of time.

[JaromirSeps]

Hi guys, the days of planning are not counted, even if it may seem so ..

[SomeClown]

And even after it hit the iceberg, it took the Titanic 3 hours to sink.  Just because you're still above water does not mean the boat is not sinking.

[dusherwo]

I wonder if JaromirSeps was referring to:
http://www.anaplan.com/
???

[Rutulian]

Oh my.  That's nice.  Very nice.

I'd see the big problem with cloud-based planning being willingness or ability to let such sensitive data outside the organisation.  Great for SME's, but larger enterprises might prefer to run on an internal cloud.  That said... salesforce.com did it.

Cheers,
Alexis

[JaromirSeps]

didn't mean Anaplan nor http://www.gooddata.com/ :)
.. sorry, can't tell more under NDA ..
cheers Jaromir

[cogplan]

 :D Nice to see so many people interacting and sharing their views on EP and TM1.
        Lets not mix and match with AnaPlan. I do feel organisations will not be willing to share such high senstiive data in Cloud as Cloud Computing is still in the beginners phase . Lets wait and watch.

             Thanks a lot for your suggestions. What would be suggestion for me who's 4 year old in this space . what can you suggest me in enhancing my skills both functionally and technically (Looks like there's enough food in my platter ,currently looking at the trends.. )
Its slightly a personal question but for a person of less experience its quite common to move on to other strreams or technologies. Hence I would like to hear from the Biggies in this field.

[nedcpa]

CogPlan,
Did you think about looking outside Cognos? Maybe Hyperion or SAP tools.

If you want Cognos Analyst type of model building experience, then go with AnaPlan. AnaPlan is a hosted solution and it combines both Analyst and Contributor in one place. Check this one out http://vimeo.com/18675160.

-Ned

[ykud]

I continue to be surprised that TM/1 is pushed, as it seems to me the best technical solution to all of this would be to slap TM/1's calc engine behind Contributor's UI.  Not just the mungey Contrib-client-for-TM/1, but sticking that calculation engine into the Job Servers and having something that turns Contributor XML into TM/1 models.  This would probably take a few years to develop.  IBM/Cognos has had Applix for a few years now.  I'm sure the internal politics of killing 2 products to make 1 better than either (not to mention migration paths for existing customers) are unfun to navigate, though.

What interests me from a far-too-technical-to-be-useful perspective is whether or not such a product would also use the TM/1 engine in the client, or if J would live on there - I think TM/1 is C++ based, whereas EP's history is in functional languages like APL and J... if you were into intellectual self-abuse you might try to implement a functional domain language on top of TM/1's engine.  It'd be a big roll of the dice, but the ability to iterate in future would be much quicker than a one-shot EP to TM/1 conversion and maybe we'd get the ability to write custom BiFs again!

Since I've spent some time toying with this thoughts as well:

Main TM1 problem currently (and in past) is a very pessimistic locking mechanism, which makes multi-user applications a very hard design problem. There were no such problems with EP, so people coming from this side of the wall just don't expect it. Happily, 100s of planning users are a rare requirement.
On the other side, EP model size constraints are actually a nightmare as well, especially for people on top of submission ladder, who need to see 'whole picture'.

Blending the two -- EP for data gathering 'bottom-up' and TM1 for data analysis and 'top-down' seemed a very nice idea. We even did a rather interesting PoC with EP + TM1 integrated just for this scenario.

Too bad that it's easier to choose 'discard one' strategy for IBM.

Techie2Techie: It relatively easy to emulate J via TM1 API, I just don't see the point in it. If EP had an open API as well, you could try and "sew" them together, but EP's a closed box.

- Yuri.

[cogplan]

Hi Techies
           Thanks a lot for all your views and experiences. I do feel the same , there needs a blending product which binds useful features at both ends, that will make TM1 - EP tied product  a super success.

          As one techie mentioned about Hyperion n SAP. I have heard good reviews about Hyperion Products spec on HFM . How do you compare it with TM1.
       
       Deviating from Original post ,
       In comparision which is better Cognos EP, TM1 or Hyperion planning products. in terms of Return on investment for the clients and also the pricing. As my client is starting to validate for new options. I would love to hear from seniors in this field to upgrade my skills and be marketeable and also helping my client vision into a right track.

      I have close to 5 yrs on Cognos BI and Planning, hence I feel Tm1 would be better option for me. Hence throw some light when you people are free.

Thanks Once again to everyone.

Regards
CogPlan