Skip to Content

For many years, Business Objects has been at the leading edge of business intelligence software.

There are two main reasons behind this; the software and its evangelists.

To put it bluntly, Bernard Liautaud had a vision of how business intelligence software could work. He was right. At its most simplistic, you need an administrative layer, a semantic layer and a presentation/reporting layer. This manifested itself as Supervisor, Designer and Reporter. Very little changed for many years because Bernard had gotten it so right. The major change was from v3 to v4; as we moved from Windows 3.11 to Windows 95, Designer introduced an interface where we could drag joins rather than hand-coding them. That was it. Contexts and aliases have been our friends for longer than my son has been alive and he’s a full-on teenager who laughs at cassette tapes and barely remembers VHS. The absolute perfection of what is now know as the UDT absolutely must be kept intact. While you have all the functionality of data federation built into IDT, you have a tool in UDT that you under-estimate at your peril. A well-designed data warehouse coupled with a well-designed universe is a wonderful implementation; a joy to behold; a pleasure to work with.

The end tool matters little by comparison. As much as I was a fan of Deski, having know it for so long, Webi has evolved to take nearly everything from it to render it obsolete. IDT cannot do that with UDT. It complicates it, adding superfluous layers. Yes, I understand why it’s there and that DF wasn’t a great tool on its own but similarly, SAP, know this; there are many out there who are masters of the universe. They are not SAP BW customers but do not neglect them. At the very least, accommodate them; they know your tools better than you do at the moment and should be seen as a rich resource of market share and R&D.

Many reading this who are from a Bex/SAP BW background simply won’t get what I’m talking about. Those who are from a traditional BO over a database background will do. The IDT will prove useful when reporting over databases but only if there is no desire in that organisation for a reporting database, if there is no desire for trending and historical reporting. So, SAP, consider the future roadmap and consider how you change the semantic layer tool. Consider not changing it – you’ve had to do very little with it so far and that’s because Bernard got it so right to begin with.

To report this post you need to login first.

6 Comments

You must be Logged on to comment or reply to a post.

    1. Mark Prosser Post author

      Tammy, I can’t see anything in that link that relates to the semantic layer’s future in the Q&A section. Is there anything in particular that you had in mind?

      (0) 
  1. James Oswald

    Ignore the sign, my friend. Tilt on.

    /wp-content/uploads/2015/05/no_tilting_at_windmills_707298.jpg

    I actually like the IDT, but I get where you are coming from. I’m not entirely sure why they can’t at least create the same type of file, so that you could create/consume a universe in either tool and have it work with whatever friend end tool you like.

    I will say, however, that in terms of semantic layers, HANA Views are where it is at…

    (0) 
    1. Mark Prosser Post author

      James, your reply about HANA views sums up the attitude of SAP that experienced BO people are seeing – if in doubt, throw money at it. Not having a go at you personally but Business Objects was always a tool for all sizes of company. The original Supervisor/Designer/Reporter product could be run off a PC with MS Access as the repository if you particularly wanted. My concern is that SAP are turning their back on BO’s traditional market and it feels like if you’re not buying HANA or BW then they’re not interested. If someone tidied up that original stack and brought it into the modern age, badged it as BO Lite, there would still be one heck of a market for it. No webi, no SIA, just a Supervisor-driven repository looking after the administration and Deski being the reporting tool. Put at the right price-point, it could dominate the cheaper end of the reporting market. If companies grow, then they could convert their universes and reports to unx/webi as per the current migrations that take place.

      In terms of the IDT, I don’t see the need for a separate federation language/layer when putting a universe over a data warehouse; you should simply be able to leverage the warehouse and its proprietary SQL language, be that SQL Server, Oracle, Teradata, Netezza, etc. As you say, though, if they can find some way of integrating the two so that you can just have one tool with or without federation then that may be a suitable end goal for all.

      (0) 
  2. Henry Banks

    Hi,

    I hear you Mark. It’s undeniable – our development priorities have shifted to prioritise the SAP install-base, over other ‘Agnostic’ (non-SAP) datasources.

    Another divergence, which unsettles some, is the focus towards “on demand” (cloud-hosted) applications, as opposed to on-premise solutions.

    It’s a tough pinch, but: SAP is now a Cloud company, running on HANA. that’s the strategy.

    However, and i hope to provide some assurance here: there are still some influencial BOE stalwarts here in SAP who champion the middle-ground, see value in servicing the SMEs, working with the non-SAP install-bases, with the on-premise customers etc. We do what we can to represent you!

    I wonder if this Go To Market survey from Merlijn could be a good vehicle for your feedback? http://scn.sap.com/community/bi-platform/blog/2015/05/20/share-your-insights-for-the-future-of-bi-complete-the-barc-bi-survey-2015

    Kind regards,

    Henry

    (0) 
    1. Mark Prosser Post author

      Henry,

      Thanks for the survey but it’s not possible to complete it based in the department that I am without guessing. There’s no “don’t know” option on one of the pages. So I’ve guessed 🙂

      (0) 

Leave a Reply