(Not logged on) | Register | Log On

You can subscribe to this discussion group using an RSS feed reader. The Joel on Software Discussion Group

A place to discuss Joel on Software

This community works best when people use their real names. Please register for a free account.

Other Groups:
Joel on Software
Business of Software
Design of Software (CLOSED)
.NET Questions (CLOSED)
TechInterview.org
CityDesk
FogBugz
Fog Creek Copilot


The Old Forum


Your hosts:
Albert D. Kallal
Li-Fan Chen
Stephen Jones

The problem isn't the suits...

Suits are so pre-Y2K in banks! Buying and wearing suits would be the least of problems with banks. But, Joel knew this and he has also identified one of the major problems with being a developer at Banks and Corporate IT ...

Your a cost center! And as a result of being a cost center, short cuts are taken. This could be poor requirements, too few developers, not real QA, etc. I could go on and on about things that I've seen. As a result, more and more the decision is made to purchase "Vendor" Products. Well, the reality is that the vendor solutions are NOT out-of-the-box solution. And if you would do it right, then I have no doubt that you can build a better in-house system, then the vendor crap.

So why do I do it? (A) Money. I make more then folks at Software Development companies. (B) Financial Software Development Companies aren't that good. I would love to find one that really had there act together. Unfortunately, after the vendor apps I've seen, I think 80% of them are crap!
Patrick From An IBank Send private email
Monday, November 02, 2009
 
 
My understanding (based on what I read on this and other boards, I've never worked in a bank) is that the real problems are the long hours, the conflicting needs of various people that you're accountable to, and the ultra-high-pressure environment.  The unclear requirements, poor QA, etc. seem to be par for the course!

Suits, on the other hand, aren't that big a deal to me.  The problem comes when all you have is ill-fitting crap that came off the rack at JCPenney.  As I've mentioned in dress code discussions, nobody would gripe about having to wear a suit if the company helped the employees get some good ones that fit properly.
Justice Walker Send private email
Monday, November 02, 2009
 
 
Before you go off on a tirade about the quality of other people's products, I suggest you check your spelling a grammar a little more carefully.    The rant become ironic when you have a high error density...

For examples of successful "financial software" companies:

http://www.mondovisione.com/index.cfm?section=news&action=detail&id=71883

http://www.rti.com/industries/financial-services.html

there are many more.

I assume you are ranting about something for a reason, but your post let me scratching my head...
Tim J Send private email
Monday, November 02, 2009
 
 
@Patrick

I work in a financial company in London -- my work hours are 8.30-18.00. It's certainly less than Google.

As for the other things, it all depends on the quality of your manager.
quant dev Send private email
Monday, November 02, 2009
 
 
I'm the lead developer at a hedge fund in Denver.  Our hours are 7-3 to match US market hours, we discourage overtime.  The code base is the best I've ever worked with (I started the project, maybe my bias is showing), we are test driven and serious about code reviews. 

I love working in finance: the problems are hard, the compensation is excellent and when you build trading systems you get instant feedback from the market if your system is working or not.
Ted Graham Send private email
Monday, November 02, 2009
 
 
Look at it this way.

Why would anyone need to write good software, when they can get the government to give them trillions of dollars in bailout?

The reason that financial companies have lousy software is that it isn't really a free market.  The value of their government-granted perks is much greater than the value of writing good software.  Therefore, why bother writing good software?

Even if I could write better financial software than Goldman Sachs, I won't be able to profit more than them unless I have billions of dollars in capital and purchase Congress and the Treasury department.
FSK Send private email
Monday, November 02, 2009
 
 
Tangent: are Goldman Sachs winning because they have better information systems than the others?

(Asking because some of the last few JoSers seem to work in the industry and might have considered opinions)
Architecture Astronaut Send private email
Tuesday, November 03, 2009
 
 
Sure, I'm recently experiencing bringing yet another vendor app in-house. Let's spend 2 million and have 10 people supporting/developing it. There was an in-house system. They were asking for 7 people to refactor over 1 year. The vendor app was chosen - wrong!

But, it's just not this. It's a bad trend I'm seeing. We're becoming integrators, not developers.

As for the hours. Mine are pretty reasonable, most week 45 hours. Sure, the past few weeks around 50-55.
But, in the past, I've done 60 hours for about a year and a few weekends. While the bonus is nice, it didn't compensate for this.

I'm whining!
Patrick From An IBank Send private email
Tuesday, November 03, 2009
 
 
Patrick

The problem is that companies who rely heavily on software, but DO NOT have software people in a decision-making role, tend to make seriously flawed long-term mistakes. 

Your industry is not alone.  Bean counters, salesmen, MBA types, etc - all seem to lack the ability to grasp some things that many developers understand. 

Oh well, there are very few perfect employers for our line of work.
Tim J Send private email
Tuesday, November 03, 2009
 
 
And while I complain, let me say that every once in a blue moon, you do find an internal group that has their act together and is doing good software development.
Patrick From An IBank Send private email
Tuesday, November 03, 2009
 
 
"Tangent: are Goldman Sachs winning because they have better information systems than the others?"

Yes. Goldmans have a more far sighted approach to IT than their competitors, and have done for many years. This is pretty evident in the breadth and maturity of some of the stuff they've built. They've generally had a build over buy mentality, and while of course like any big company there are some horrible legacy systems lying around, the general patterns is that some very smart people got a lot of stuff right from the beginning. If you look at their risk systems, order management platforms, algo engines, crossing engines, tick databases etc it's all pretty good stuff.

Because of the investment in technology over many years, it takes time for competitors to close the gap.

I work in the industry (front office algo & quant type role) and the hours are reasonable, the pay outstanding, and I get to work with some very talented people. Yes there is an enormous amount of pressure to deliver stuff fast, but generally the mentality of front office developers is to release fast and often, and yes sometimes that comes at the expense of quality. Some of our back office guys are quite religious about methodologies, agile, test driven development etc - but what you'll usually find is that in the front office, 'getting stuff done' is king. We don't have a lot of time to quibble over which version of a unit test framework offers that great feature, and while some developers don't like this approach (they consider it accruing technical debt which you'll have to pay back later), somehow we still manage to crank out quality stuff with an absolute minimum of bugs and maximum reliability.

As compared to working for a proper software development house producing boxed software - I would rather work at the i-bank. For me (unlike some other programmers I know) software development is not an end in itself - I like building stuff that is actually used and where I get immediate feedback on how it performs - which I wouldn't get in producing software for the general public.
Lincoln Marr Send private email
Wednesday, November 04, 2009
 
 
""Tangent: are Goldman Sachs winning because they have better information systems than the others?"

Can't ... resist... http://www.correntewire.com/goldman_sachs_are_scum
Eugene Kaganovich Send private email
Wednesday, November 04, 2009
 
 
Very timely question, as I have 3 large blog posts brewing about software development and the financial industry. I've worked in investment banking for 30 years.

The bottom line is that the (very good) money available means that good developers, analysts, and project managers can be bought. And it's the quality of these people that determine the quality of the software, not the software development process.

So what's *really* important is the quality of the hiring process. If you get that right (for the value of "right" that suits your team's culture), then you've got a large head start over companies that don't get it right. Many teams at Golden Sacks get that right.
Mark Pearce Send private email
Thursday, November 05, 2009
 
 
Mark - I'm looking forward to your posts.

I've heard mixed things about Goldman. Yes, pays the best. Also, a "Get it done" Mentality in the front-office. The company very much has the attitude of "We're Goldman Sachs" and a lot is about playing politics. From what I've heard, they do whatever it takes for Goldman.

Personally, the only experience I ever had with them was after working for about 5-6 years, I was applying for a position there. They wanted college and HS transcripts. Forget about it!

Anyways, I'm feeling a little bit better today. More vendor bugs have come out. So they know what about decision it was.
Patrick From An IBank Send private email
Thursday, November 05, 2009
 
 
The "bulge bracket" banks are indeed about politics. There is good politics and bad politics. It's good to learn how to play the former, and sometimes necessary to play the latter.

If you're a developer, you tend to be shielded from much of the direct politics, although obviously you feel the effects. But often it's about making the decision right, rather than making the right decision. I think that I'm actually a better developer after having gone through the politics.
Mark Pearce Send private email
Thursday, November 05, 2009
 
 
No offense, but why would any developer want to be working in this field right now???

This may be a gross generalization, but are we not in a period where the banks, the lendors, the hedge fund, mutuals, and Wall Street in general proven to be irresponsible, mismanaged industries? I mean, how much regulation is coming your way? All those hedge funds and derivative calculators and regression analysis r-squared BS tools you guys wrote are responsible for much of the BS and mess we are in now.

In stead of working in a field that moves stock trades and customers around on a computer screen for the rich and financial elite, why not work in a field that has some real growth and meaning....hmmm.....like software? What comes around goes around I guess.
Website Photo Gallery Guy Send private email
Saturday, November 07, 2009
 
 
"No offense, but why would any developer want to be working in this field right now???"

Because the pay is good and the job interesting.
quant dev Send private email
Saturday, November 07, 2009
 
 
>> This may be a gross generalization, but are we not in a period where the banks, the lendors, the hedge fund, mutuals, and Wall Street in general proven to be irresponsible, mismanaged industries? <<

Yes, you're making a gross over-generalisation. I know this may seem ambiguous to basement-dwelling cubicle mutants who live, breathe, and shit software, but real life is nuanced and subtle and requires a kind of mental hygiene.

>> I mean, how much regulation is coming your way? <<

More regulation means more software that needs to be designed and built :-)

>> In stead of working in a field that moves stock trades and customers around on a computer screen for the rich and financial elite, why not work in a field that has some real growth and meaning....hmmm.....like software? <<

Gordon Gekko stated it best: "Greed, in all of its forms; greed for life, for money, for love, knowledge has marked the upward surge of mankind."
Mark Pearce Send private email
Sunday, November 08, 2009
 
 
Patrick, my advice is that you find an i-bank (or move to a team) where "build" is just as important as "buy". That may mean moving nearer to the front-office.
Mark Pearce Send private email
Sunday, November 08, 2009
 
 
"So what's *really* important is the quality of the hiring process"

@Mark Pearce: spot on. That's exactly what we've found, and over time we've refined our hiring process a fair bit having been burnt early on by a few bad hires. Any hires we take on now tend to hit the ground running and do well.
Lincoln Marr Send private email
Sunday, November 08, 2009
 
 
RE:  "No offense, but why would any developer want to be working in this field right now???"

The field has a quite a number of interesting and challenging aspects to it (we're not talking about P&L and ledger software...)

There is also a LOT of money on the table at financial institutions.  They are happy to pay big $ if you deliver results that translate to profits. 

Not everyone wants to make shrink-wrap software or the latest online "community" web page.
Tim J Send private email
Sunday, November 08, 2009
 
 
>  I mean, how much regulation is coming your way?
But, who says they will be followed?

Sarbanes-Oaxley was supposed to tighten controls. But, it hasn't in all cases.

Great advise from Mark! I don't always agree that moving to the front-office will make a difference. I've certainly see quite a bit of buying trading systems. Actually, now we support an "Execution" desk and it'a vendor app.

In some ways, I've had better luck with the middle office and Workflow type systems.

But, I do agree that going to a different team/group is a good idea. I interviewed with a different team/group, but probably didn't get it. They were looking for some C++ and I don't have it.
Patrick From An IBank Send private email
Monday, November 09, 2009
 
 

This topic is archived. No further replies will be accepted.

Other recent topics Other recent topics
 
Powered by FogBugz