Don't be so harsh on all of IBM, its like hundreds of companies under one banner. Not all of them have the power to influence other products and often times end up stuck with it and cannot move on to other tools (i.e git).
More often than not, customers have higher priority on bugs than other teams dog-fooding the product. Its insane.
I think you mean don't be so harsh on the people on the ground at IBM. If the issue is IBM's internal structure is fractured and disorganized then I think it's perfectly fair to blame the company.
The do this so different divisions can lean on each other and benefit from the IBM brand name. If they want the benefits of this shared branding then the deserve the downsides from it too.
I've noticed many companies recently thinking they can have it both ways (samsung and sony come to mind). Live by the brand, die by the brand!
My own anecdote -- we had a need for a GIS/charting/mapping system. IBM responded to the RFP as did many others.
IBM proposed an entire office automation system with a minor charting system hanging off the side somewhere. Most of the components were unfamiliar to everyone. Though maybe there was Wordperfect or Lotus Notes in there. I remember meetings with IBM always had whole bunch of shiny suits whereas more focussed proposals just had one or two people visit.
After one or two rounds of pointless meetings our fairly technical director told them to go fuck off and never come back.
The worst thing though that was despite paying an inordinate amount of money to license this software each year and receive support for it they would not fix a bug we found in their mapping modules.
Specifically it was their use of float for X and Y coordinates in the IlvPoint object that you pretty much had to use to make a map work with their tech. In certain parts of the world there is an extra digit on the Y coordinate (projections) and the float they insisted on using couldn't store the whole value, meaning they would just round it up or down causing inconsistencies between our representation of a client's dataand the client's representation. That was an absolute no-go for us and the workarounds IBM suggested to us were absurd and bordering on insulting honestly.
We had a lot of trouble with our affected clients because of this and all we could tell them was that it was an IBM issue.
The absolutely refused to fix their product and eventually sold it to Rogue Wave, washing their hands of it. By that point though we had already decided on a few alternatives lest we start to lose customers (who were essentially massive utility companies from around the globe).
Fuck IBM.
We are starting a project and even with FTE's within the company, "consultant" has convinced management to hire IBM resources because of cost. The high sounding reason within the presentation twisted it to - FTE's are not expert in this tech so they won't be able to do an efficient job.
But it is always that the end resources from IBM are even more incompetent. They end up delivering a product which is terrible. Even the final handovers and KTs are done terribly.
In the end, management has no option but to continue hiring IBM (or any service business) to support the shoddily done project.
This doesn't mean all implementations from a services business is terrible. It all comes down to cost and ability to spend. If you pay peanuts, you get monkeys - this is aptly true.
To answer your last question - Yes there are different entities within IBM. There is GBS which takes up most of the consulting and up-selling IBM products and services. Development for example Watson is done by IBM Labs. A list can be found here: http://www-07.ibm.com/in/careers/businessunits.html
At an old employer who I'm no longer with, I asked why there's a part of the application in Java (the rest has always been in .net). They told me that an army of IBM consultants made that decision and now there's a part of the code in Java and it sticks out like a sore thumb.
But doesn't the word get around ? does nobody in most enterprizes think long-term when investing ?
IBM is optimized in enterprise (including government) sales. Product is not quite irrelevant, but it's a specialized enough market that often mediocre substantive quality does quite well, because the only competition that matters are the others that are optimized for that market.
Wasn't there a saying that no one got fired for buying IBM.
I've seen it many times.
Misinterpretation is a recursive problem. The best description - by far - of this process is "How a plan becomes policy" (aka, "How Shit Happens").
As the staff back away slowly from the creep, that just makes more space for IBM, and they sell progressively more service hours. It's pretty much their MO.
Broadly, yes, IBM is run as various different divisions, and the divisions start very high up the organisation, only a couple of levels from the top. They rarely interact in my experience (as a software guy there for some years in the past)
I think it expected to be installed in /usr/local and company procedures demanded that it was installed in /opt (or the other way around).
For this it would consequently misplace some permissions on some of the files causing a later stage in the installer to fail and bork the installation.
I bumped into her last year at Ansible Fest London and she was still angry about it 17 years later!
And that's what Oracle does to your soul :)
Edit: that N class and Oracle NEVER got deployed in the end. It just sat there unused, licenses weren't renewed, it got sheepishly powered down after a few months to save electricity and it got chucked in a skip after 5 years. About £200k of expenditure down the pan.
I guess you could easily lost even more money. Good thing they stopped it at £200k.
The whole platform got replaced with SQL Server on a clone PC server in the end which cost less than £10k. No fan of Microsoft but there was no competition. And the tooling was far better!
Legal guys said it wasn't worth suing them because we'd have to pay up then or lose the support contract.
> However, I think you'll see this shift towards a developer-focused mindset happening more-and-more within IBM over the next few years.
means? In a discussion about not trusting IBM's marketing, you can't simply say "I hear you and it will get better, trust me."
I don't work in the same role so I only know things from a 10,000 ft view. But a lot of orgs that IBM works with have inane bureaucratic structures with their in house devs pretty much at the bottom of the food chain, or in an "IT department". All decisions on using tech tools are made by "higher ups" a lot of whom have 0 technical background. We're trying to change this culture to one where devs get to make those decisions AFAIK.
Disclosure: IBM dev, but not a Developer Advocate.
IBM was always unique, could sell you sfw, hdw, hosting, resources, and finance the whole thing. Some business as leaders never really adapt to change because management is stupid (blockbuster, radio shack, blackberry, etc.). IBM had so many chances to compete with Amazon and they failed. Watson is no more than a deflection in the marketplace that masks the real problems from within.
If you can't provide any useful comments, then nothing. What they're saying contributes absolutely nothing to the conversation. Maybe say you'll stop selling non-existent features to executives for starters? I don't appreciate HN becoming an advertising ground to name-drop IBM without any real talking points, please provide some sort of real discussion.
> We're trying to change this culture to one where devs get to make those decisions AFAIK.
Ok, but again how? Or I guess in other words, I don't believe you.
The tradition was still going at that point.
https://www.itnews.com.au/news/queenslands-ibm-ban-lives-on-...
http://www.cbc.ca/news/canada/ottawa/phoenix-pay-system-psac...
What do you mean by 'over engineered' in this context? The usual way I've seen that term used would imply 'reliable but more expensive than necessary' rather than 'unreliable'.
Think J2EE and their xml config files.
Or using hadoop and kubernetes when a sqllite db on digital ocean will do
I have seen middle management say very clearly that no one was ever fired for choosing IBM.
Is that just a notable exception or are there better alternatives out there these days for hosted CouchDB?
Not saying it's not good (I've never used it), but IBM can't really get any credit for this. They just happen to have bought a decent product and now they sell it.
I like Cloudant .. it is not too bad. In a recent project (after I left IBM), I needed to get a data tier up quickly and I actually used it for the initial POC. And yes, it was an acquisition by IBM. I was a bit worried that bluewashing would have made Cloudant crap but looks like they have managed to avoid it. The caveat I'd suggest is look closely at the pricing model. If you want QoS and scale, be careful of the pricing model. For small projects and projects that need fast completion (what project today doesn't?), it is actually very nice. You asked for caveats .. I think some of the language bindings were a bit out of date .. but since it uses REST calls, it wasn't too hard to bypass.
It's really easy (largely thanks to Cloudant) to host your own Couch instance but it's also really affordable to let them do it.
Their tools are trash for the cost they want. The majority of their offerings are overengineered and prone to failure in a production setting.
They are pros at selling to non-technical management, all the while IT employees get stuck dealing with the aftermath.
We ended up getting stuck with the IBM version of work item management, called Jazz Team Concert. From the "source control" to the actual setup and configuration, the product was a nightmare. The worst part was when we looked at license renewal. They wanted three quarters of a million dollars for something that was essentially dog shit wrapped in cat shit.
In case you thought I might be on the fence with IBM, I'll add one final clarifying point:
Fuck IBM.