What’s happening to the next generation of the SAP ecosystem? • Hacker News Board

Re: What’s happening to the next generation of the SAP ecosystem?

#11

Post by paganel » Thu, Jun 16, 2011, 2:46 PM UTC

> Big data is sexy. Ruby on Rails is sexy.

This is really a pity, because large companies can gather lots of interesting data. For example, my wife works for a large (European) oil company, which has been in the process of installing SAP (it only took ~ 2 years, with lots of SAP consultants paid at 800-1000 euro per day ).

Anyway, being the big company that they are they have lots of gas stations spread all over the country, in the best locations possible. But nobody is doing anything with that sales data. They could write some heatmaps, make some sales predictions based on how close a gas station is to a major road, see when things change etc. There's no incentive for doing that, because it will either involve re-hiring those consultants at 800 Euro per day or trying to train the internal IT people to modify/extend the existing SAP installation (good-luck with that!)

Re: What’s happening to the next generation of the SAP ecosystem?

#12

Post by parfe » Thu, Jun 16, 2011, 2:49 PM UTC

brendino wrote: post #9

SAP's lack of openness is a huge problem, and will continue to hinder its development. Not only does the license cost millions of dollars, but then the enterprise that purchases it has to hire dozens to hundreds of expensive consultants to actually implement the behemoth. And if the company wants to actually understand how to use the system, they need to invest in expensive training sessions because any sort of publi…

>I tend to think that the lack of openness is done on purpose

I doubt it. From my experience it's from outsourcing development to too many different groups with no coordination.

I worked on interfacing with a BigName's Hour Time Tracking module which takes hours to 8 decimal places. I submit the first payroll and hear it is imported with no errors.

Two days latter I get a frantic email asking me to resubmit because the Payroll module only takes 6 decimal places and was throwing out all the submitted hours. The systems are so large the consultants themselves cannot possibly understand it, let alone provide proper documentation.

Re: What’s happening to the next generation of the SAP ecosystem?

#13

Post by bchjam » Thu, Jun 16, 2011, 3:02 PM UTC

paganel wrote: post #11

> Big data is sexy. Ruby on Rails is sexy. This is really a pity, because large companies can gather lots of interesting data. For example, my wife works for a large (European) oil company, which has been in the process of installing SAP (it only took ~ 2 years, with lots of SAP consultants paid at 800-1000 euro per day ). Anyway, being the big company that they are they have lots of gas stations spread all over the…

I think they'd need to license one of the bulkier versions of BusinessObjects to get the BI angle out of SAP. Not saying it couldn't be done in house but the mentality of using something like SAP doesn't seem to fit with the home-grown approach that well.

Re: What’s happening to the next generation of the SAP ecosystem?

#14

Post by jriddycuz » Thu, Jun 16, 2011, 3:16 PM UTC

This guy pretty much gets it. Young programmers, particularly talented ones, are going to think that anything enterprise-y is just lame. Working at a startup feels like keeping it real in a kick-ass garage band. Working at SAP feels like playing triangle in a Gamma World City's Philharmonic. Even if it pays well and affords a modicum of respect, it just don't do.

Re: What’s happening to the next generation of the SAP ecosystem?

#15

Post by bonzoesc » Thu, Jun 16, 2011, 3:28 PM UTC

blumentopf wrote: post #2

I had a customer project two years ago: Import data that came out of SAP HR-PT into a MySQL database. HR-PT is the SAP personnel time recording system. This thing generates plain text tables with each entry having a fixed-width data type which has a peculiar German name. The 21st century solution for this would be an XML file. So the impression I got is that the whole SAP codebase seems to be stuck in the 1970s. Work…

> Working with this stuff is so complex that in Germany, it's an often-told inside joke in the IT industry that some small and medium sized businesses actually became insolvent by moving to an SAP-based system.

I've heard of this almost happening; a malfunctioning label printer can cause an entire plant to shut down for weeks while consultants work to find the problem.

Re: What’s happening to the next generation of the SAP ecosystem?

#16

Post by ultrasaurus » Thu, Jun 16, 2011, 3:32 PM UTC

parfe wrote: post #12

brendino wrote: post #9

SAP's lack of openness is a huge problem, and will continue to hinder its development. Not only does the license cost millions of dollars, but then the enterprise that purchases it has to hire dozens to hundreds of expensive consultants to actually implement the behemoth. And if the company wants to actually understand how to use the system, they need to invest in expensive training sessions because any sort of publi…

>I tend to think that the lack of openness is done on purpose I doubt it. From my experience it's from outsourcing development to too many different groups with no coordination. I worked on interfacing with a BigName's Hour Time Tracking module which takes hours to 8 decimal places. I submit the first payroll and hear it is imported with no errors. Two days latter I get a frantic email asking me to resubmit because t…

True, I've worked at OtherBigName, and whereas we did make as much money off training as software, the documentation mostly suffered because writing good and useful documentation of spaghetti code is really really hard.

Re: What’s happening to the next generation of the SAP ecosystem?

#17

Post by c00p3r » Thu, Jun 16, 2011, 3:49 PM UTC

paganel wrote: post #11

> Big data is sexy. Ruby on Rails is sexy. This is really a pity, because large companies can gather lots of interesting data. For example, my wife works for a large (European) oil company, which has been in the process of installing SAP (it only took ~ 2 years, with lots of SAP consultants paid at 800-1000 euro per day ). Anyway, being the big company that they are they have lots of gas stations spread all over the…

Very common, standard story. Huge amount of time and money being wasted just because SAP's sales people successfully deluded top executives by slogans like 'best run business runs SAP' and the like.

The most of that data analyses could be done even with open-source storage and tools and frameworks like Rails. Btw, BusinessObjects itself is just a bunch of open (apache, and the like) Java packages glued together under tomcat. Of course, there are a lot of custom code, 40% in my estimation.

There were a lot of attempts to create and which is more important to push to the market various open source ERPs. They are failures because there is no demand for ERP without big money and big name behind it, just that managers could be able to hide their asses behind the papers with big names and long words.

There is demand for a relatively low cost, specialized in-house solutions based on the same open source technologies as facebook or google, but that is very difficult to execute without spending a lot of money on research, business analysis and collaboration with each particular enterprise.

Re: What’s happening to the next generation of the SAP ecosystem?

#18

Post by brendino » Thu, Jun 16, 2011, 4:26 PM UTC

parfe wrote: post #12

brendino wrote: post #9

SAP's lack of openness is a huge problem, and will continue to hinder its development. Not only does the license cost millions of dollars, but then the enterprise that purchases it has to hire dozens to hundreds of expensive consultants to actually implement the behemoth. And if the company wants to actually understand how to use the system, they need to invest in expensive training sessions because any sort of publi…

>I tend to think that the lack of openness is done on purpose I doubt it. From my experience it's from outsourcing development to too many different groups with no coordination. I worked on interfacing with a BigName's Hour Time Tracking module which takes hours to 8 decimal places. I submit the first payroll and hear it is imported with no errors. Two days latter I get a frantic email asking me to resubmit because t…

That's a good point - enterprise software is meant to be customized to a business, so maintaining a strong repository of documentation is difficult, if not impossible.

However, I've noticed that even some of the core functionality is pretty light on documentation as well (and it is typically written in German-English).

Re: What’s happening to the next generation of the SAP ecosystem?

#19

Post by pasbesoin » Thu, Jun 16, 2011, 4:44 PM UTC

blumentopf wrote: post #2

I had a customer project two years ago: Import data that came out of SAP HR-PT into a MySQL database. HR-PT is the SAP personnel time recording system. This thing generates plain text tables with each entry having a fixed-width data type which has a peculiar German name. The 21st century solution for this would be an XML file. So the impression I got is that the whole SAP codebase seems to be stuck in the 1970s. Work…

I've done somewhat similar work. I found it best ultimately to pull individual transactions -- as a text/report dump -- and then work them up in a combination of Excel with hooked in regular expression support and Perl and whatever else (i.e. for a front end interface). Excel provided a convenient interface for necessary manual sanitation. Perl provided enough speed along with flexibility and further pattern matching. Other stuff made the results accessible to accounting staff.

For my purposes, SAP was a giant data entry system. And a crappy one, at that, as one of the most important fields for my work was not a field, but one of an arbitrary number of entries glommed together in a free text field where only the maintenance -- or not -- of data entry conventions by the many individuals creating records provided any degree of sanity.

Thank goodness for regular expressions!

Separately, it's not just the "sexiness" of the problem at hand. It's that in typical SAP environments, the level of bureaucracy to get anything done is overwhelming. Your technical problems are far and away secondary; for example, just getting what you need on a report, any report, in a format you can at least begin with, is a six month ordeal.

Having your hands on the pulse of billion dollar divisions is kind of cool. The six or eight layers of wankers you have to go through, to accomplish this -- not so much so.

Re: What’s happening to the next generation of the SAP ecosystem?

#20

Post by mbesto » Thu, Jun 16, 2011, 4:48 PM UTC

jriddycuz wrote: post #14

This guy pretty much gets it. Young programmers, particularly talented ones, are going to think that anything enterprise-y is just lame . Working at a startup feels like keeping it real in a kick-ass garage band. Working at SAP feels like playing triangle in a Gamma World City's Philharmonic. Even if it pays well and affords a modicum of respect, it just don't do.

It does pay well, which is why I'm often surprised there's not more demand from the younger generation. For example starting salary for a SAP consultant at a big firm starts at about $50k with some nice cooshy bonuses on top of that. Very competitive with that of banks actually. But bankers have the "promise" of making several folds of that if they just stick it out. I don't think the younger generation sees that promise in the enterprise world - even though I know many SAP in their 30's and 40's who are making more than enough to support a family.

On the flip side there