- Feature Articles
- CodeSOD
- Error'd
-
Forums
-
Other Articles
- Random Article
- Alex's Soapbox
- Announcements
- Best of…
- Best of Email
- Best of the Sidebar
- Bring Your Own Code
- Coded Smorgasbord
- Mandatory Fun Day
- Off Topic
- Representative Line
- News Roundup
- Editor's Soapbox
- Software on the Rocks
- Souvenir Potpourri
- Sponsor Post
- Tales from the Interview
- The Daily WTF: Live
- Virtudyne
Admin
You nailed buddy! thanks for the post.
Admin
Sorry, but this is the kind of post I just don't want to read here. Honestly, I don't care about your opinions on the matter, and I think that you are dead wrong about them anyway. Let's get back to laughing at poorly designed code and the curious behavior it causes, along with the personal stories from those in IT that have to deal with the severe WTF.
Admin
It's called a 401(k). (Or sometimes an IRA). Assuming for a moment that you do the smart thing and manage to look around for the better job before leaving the original one, there's no reason you can't contribute plenty.
Your present is already very much dependent on your company. Why put all your eggs in one basket and make your future dependent too?
Admin
heh its funny how the company I work for is nearly the polar opposite of this article. Even the idea of me leaving is skated around and ignored, i got told once that the company was like the Hotel California - "You can checkout any time you like, But you can never leave!"
Every time i suggest certain things like better documentation in the event that I do leave, which also gets skated around and ignored.
Pay rises are nill, promised benefits are never delivered.
This has lead to me becoming less and less motivated with my job, I only learn new technology's so i can do my job in a fraction of the time and read sites like this in the rest rather than doubling my productivity.
I feel just like the guy from office space, "I only do enough work to not get fired" but im too lazy to do anything about it because i dont see much hope in any other company around.
Needless to say, looking at changing carrers.
Admin
Up or Out is also a common strategy in the Army. One part of the justification is that it provides a large reserve officer core, who can be recalled to officer a civilian army in time of conflict. It also means that junior officers are young and fit, and solves the succession problem.
But it has the effect that normally all officers are inexperienced at the level at which they are serving.
Admin
Rubbish, there are all kinds of people staying or leaving and have absolutely nothing to do with talent.
Admin
I like how the article says "This growth continues exponentially", while the graph is obviously far from exponential.
Admin
a REALLY good article
two points:
#1 often, there are not enough skilled workers to go around - as in: employees leave, but it is very difficult to find others with even minimal skill - so retention is MUCH MORE IMPORTANT in these circumstances
#2 "no one will miss him when he’s inevitably let go" is the sad truth of todays world where making money is more important than social responsibility - while at the same time we are facing a 80%-20% distribution of talent in people. what are the 80% of common/typical people with no exceptional talents to do? i think it is acceptable for companies (except the small and the startups - in the begining only) to retain some of these people and help make world a better place for all, for a small share of the profit.
again, a really WICKED article. we need (expect) more like these, from now on :)
Admin
The article is provocative. But...
Minor point: "On the first minute of the first day, an employee’s value is effectively zero." Uhhh... no.
It's negative, because the new hire is taking others' time, while contributing nothing.
It seems to be generally accepted that break-even (contributing as much as you cost) occurs between six months and two years into the job.
Corrective: Shift the curve down, and then integrate. Conclusion: it's worth spending a lot to retain able employees until their value descends back to break-even (which it may never do).
Major point: If I were working in this regime I would feel no desire whatever to write clear code, or to document properly -- in fact, the reverse. I'm unlikely to see the code again. I'm not going to help potential competitors. Let them figure eveything out the hard way. I had to. Applied nationally, this is a recipe for a huge loss of productivity. (That's good for us here in I***a, though.)
And finally, WTF?! This site is for real people, not management drones. Begone, guest-posting jellybrain! If you want to post again, it'd better be another clbuttic, like "The Brillant Paula Bean".
Admin
Wow. This is a perfect explanation of my workplace, why people are quitting and why I want to get away as soon as I can.
I have now been working three years on the same monstrosity of application. I have removed some major WTFs programmed by very senior people and written some parts that suck a bit less myself. There are still many parts of the application and our development processes that suck. I know how to fix them. Instead I am busy adding new features (no, not writing any real functionality but writing wrappers around awful third party modules). Now, the "up" Alex has been describing would not be a raise to management but instead a position where I could fix the suckiness.
The application itself is written in a restricted subset of C. Any skills I learn here have no direct application in the real world. Of course, any real world programming is also useless here. Now the product is being slowly phased out so there is no space for any newcomers to prove their skills or learn anything useful.
Of course the pay is poor, too, but I do not think any amount of money would make me want to stay (except if my time would be so expensive that the company could not afford me going through pointless processes and bureaucracy).
Admin
Meta-comment: Well, advice, really. Peer-reviewed psychological experiments have established that, as a rule, under-performers significantly overestimate their own performance, while top performers significantly underestimate theirs.
Think long and hard before claiming membership of the latter group. What's your score on SPOJ?
Admin
Sorry, but this is the kind of comment I just don't want to read here. Honestly, I don't care about your opinions on the matter, and I think that you are dead wrong about them anyway. Let's get back to laughing at poorly designed code and the curious behavior it causes, along with the personal stories from those in IT that have to deal with the severe WTF.
Admin
While this sounds all nice and trendy and resonates with the fast-paced 21st century blah-blah you know the story, it is simply not true.
Ambition to improve your skills does not necessarily imply "institutional ambition". People who set out to improve their skills every day are often labelled complacent if they are not loud enough. On the other hand, loud people with no talent and knowledge whatsoever will be recognized as ambitious if they expose themselves often enough.
In the context of the theory this means that while skilled people do tend to move on quicker, the reverse is not true, perceived ambition and skill don't go hand in hand.
Shame, because otherwise it would be a well thought-out post.
Admin
There are essentially within IT two groups: those that build things and move on (good code or shitty code doesn't matter) and those that implement and maintain stuff so businesses are able to use the app during it's lifecycle.
I'm now about 15 years in IT as an IT Infra designer/architect and I'm working for a DataCenter which enables a lot of Financial Institutions. Thus second category: keeping everything controlled, stable and secure so everybody get's their salary every month or week.
I can tell you: the least thing we want is a high turnover. Why? Because it takes relatively a lot of time to be able to entrust a novice with the responsibility for handling Business Critical systems (they usually start in a Development environment, and then move up the ladder when they have shown they have matured and can handle the responsibility for Production).
It's quite simple: we care less if a developer stays or moves on. There is always a new developer somewhere to be found. If he/she produces lousy code we'll make 'em move on otherwise they will get bored and move on by themselves. Fine by us.
However, running a business critical system and staying calm and collected while the world is shouting because there is a salary run for a big firm going awry: that simply takes another mentality.
Both kinds of people (developers and steady-state) are necessary during the business cycle of an app. The first do their thing and move on and here the Cravath business model could be successful if implemented correctly.
The other kind should be kept on board for as long as possible since for the second kind institutional and customer knowledge is key to be successful and here the Cravath model won't be successful since the high turnover would be detrimental to building and maintaining institutional and customer relationships.
In short: think before you leap in implementing the Cravath model and assess your organisational maturity before you do. If steady state is the most important thing: don't implement Cravath. If time-to-market is (even with shoddy software) then Cravath can add business value to your organisation.
Admin
logs are another way of writing exponents...
Admin
Oh, and just one more thing: if you fail to define "up" correctly, "up or out" is all you need to implement the Peter principle.
What you'll end up with is a set of mediocre managers who had once been good coders/analysts/architects and a bunch of undergraduate kids writing your code.
Admin
Logs are useful for transforming expressions that contain exponents into expressions that are easier to work with. If the y-axis on the graph were logarithmic then it might be valid, but clearly it is not intended to be anything other than arithmetic.
Admin
So, Alex Papadimoulis, given that "exponentially" is the wrong word to use, any chance you could replace it with less likely to inflame mathematicians' ire? Alternatively, you could just remove it, because it adds nothing useful to the meaning: "the growth continues" is just fine as it is. We don't need to know how the growth continues, just that it does.
Admin
Admin
I envy you!
I've always wanted to be an entrepreneur, but next to that, you've just described my dream job :)
Admin
We were talking about that at my old job; we reckoned some of the worst developers we had would benefit from having the first couple of rungs taken away. If we could just remove their sense of contentment based on physical well-being (via the use of a small, sound-proofed room, a chair and some rope), they might be more motivated to write good code.
I think we were ahead of our time.
Admin
don't you see the contradiction?
The company does not value him sufficiently to pay him what he can get on the open market. They are not willing to reward him for his skill.
If he's not worth that much money to your company then so be it, but don't get all bitter that he doesn't want to hang around and treat the company as a charity case.
And if he really did create a mess then the company doesn't need his skill. Alternatively his mentor did a poor job teaching him how to produce quality code.
Admin
I'm tired of pedantic jackasses who feel the need to correct everyone all the time. Are you trying out for "Pedantic Jackass Guru", Gamma?
Admin
You know, this is exactly how it is at my current job. And there is a lot of turnover, but it's designed to have a lot of turnover. We also have a lot of interns, many of whom become full time staff members (I was such a person). All of the staff are competent, so far as I can tell, almost half of them were former interns, and at least as many won't be working there by the end of the year, myself included.
Admin
I'm a quitter! Who would like to embrace me? :)
Admin
"Employee value peaks and then begins to decline" Bullocks! Only true if the employee never learns anything at the job.
If the company help train employees they would not find the workforce stagnating. But then again any company that buys into this theory is probably not capable of understanding long term growth and the benefit of a stable employee base of top tier talent.
Basically value-apex is a theory fit around the fact that most companies use mediocre talent and underpay current employees by giving below average raises. Employees are forced to move to get paid what they are worth. If the company pays well, gives good raises, and helps employees to continue develop the company gets both the long term stability and keeps all the rock-stars that "value-apex" shops lose in droves.
True story. I have had friends double salaries by quitting and then re-applying for their current positions at "value-apex" shops. Because when the company has to take an honest look at the employee they realize that with the experience the employee has gotten at the job they are worth a lot more then they were being paid. But giving everyone raises is expensive so they only give it to the employees who quit. I have also had friends whose salaries jumped 50% after hiring "assistants" because their company realize that starting salary for peers at their skill level was twice what their corporate formula allowed. And hiring an assistant for more than the senior developer is just asking to lose both.
Value-apex - a strategy for companies who don't value employees. If a company gives employees sufficient challenge and variation while paying an honest salary there is no reason to leave. I have several standing offers the moment I walk out the door. But I don't do that because my company doesn't treat me like crap. You can keep your value apex. I will keep the job I am doing for 8 years now and at which I am one of the best in my field.
Admin
It's so rare I hear or read these words without sarcasm being involved.
Admin
You are now assuming that the company actually wants employees to learn something. This is not true, because a) The time employees spend new things is time away from productive work, b) Employees with any marketable skills could find other work and leave the company. If people with new skills are required, the project is outsourced to someone who claims to have them. This of course not a good strategy for everyone, but for certain kind of companies it works too well to be changed.
Alex spends some time describing how the apex can be shifted (the parts called Shaping the Value Apex and Bringing Home Change). "Sufficient challenge and variation" and "honest salary" are among them. It is certain that you are still on the upward part of the curve and if you are lucky you can keep there until retirement.
For me, it is either low paid boredom or quitting. Guess which one I choose.
Admin
Whether you liked the article or not, it's certainly provoked a wonderful discussion.
I think what interferes with most IT staffing, no matter the management style, is company politics. I've noticed that in all the companies I've worked for the brightest minds tend to "rock the boat" by making suggestions, trying to fix problems, trying to change things. In the places I've worked, those people were hated by management and vilified and driven out if they didn't get disgusted and quit first. I even witnessed this at places that claimed to hire only the best. Until we solve the problem of politics first, I don't think we'll solve the problem of managing IT staffing.
Admin
TRWTF is that I read almost that whole thing
Admin
Thanks for this article--I have been considering my professional career direction over the last couple of weeks and many points in this article describe my situation perfectly. I think I'm going to have to bite the bullet and move on.
Admin
it seems startups are more like this - Based on people I've met in my current (and first) job, is that people who left their old companies on good terms, just know so many more people.
The notion of not staying forever is there too. I'm sure a 4-year vest makes deciding when to quit pretty easy.
Admin
The author meant that statement not to imply similarity between the knowledge or tasks of the industries, but rather that those others -- being around longer -- are more mature and have many of the hiring & retention practices he outlines in the article already in place. So he's saying, "let's be more like these industries, so we can enjoy some better workplace treatment and experiences."
Admin
So that's why countries have a foreign exchange program
Admin
If all the skilled people quit first, why are the majority of my projects "pick the pieces of program X" where the developer of said piece of shit has recently left the company?
And if they were the company's brightest stars, why is my whole department tied up migrating undocumented sales tools written in VBA on Access and Excel to something at least quarter-way supportable?
Sorry, but "the best leave early" just sounds like the excuse you come up with when you quit your last five jobs just prior to the chickens coming home to roost.
Admin
From my personal perspective it does seem to help. I spent almost 9 years with a software developer and in that time worked in 3 different departments each with different development opportunities. Had I not, I probably would've left after about 3 years.
Admin
I agree with the general gist of what you're saying, but you include just enough untrue absolutisms and blanket statements to discredit the entire piece:
When I was hired at my current company, I was precisely dating around, moving from place to place in search of the "Perfect Company" at which to grow old and retire from. Now, I would infer from your above quote that either (a) I am not considered to be one of the most talented developers, (b)you meant to have the word "Most" or "Some" at the start of the quoted sentence above, or (c) that the quote is just plain false.
As to possibility (a), I'm fairly certain that the company I work for, including the folks holding positions both above and below me would say otherwise. Also, there are a couple of things I've accomplished over the last 6 years that might make 1 or 2 other people (whom I've never met) consider me "talented".
As to option (b), well, it could be a simple mistake. Option (c) would be enough to discredit the entire article, in my opinion.
As with most statements that contain such absolute terms as "always" and "never", they are almost always never true :)
Indefinite retention is extremely difficult, but not impossible, unless you're counting death as a form of "quitting". I think you mean "100% retention is impossible; there will always be employees who quit.". After all, if indefinite retention was impossible, there'd be no such thing as retirement.
While there may be employers who have a hard time accepting that employees quit, I think most employers just don't like employees who are performing well (and who have had a significant investment placed on their training) to quit. We have a fairly high turnover here (in fact, a really high turnover right after bonuses are given out). It's talked about and dealt with as an inescapable (and undesirable) fact, but I don't think any one here has difficulty accepting that it happens.
I could go on, but I will give that this is, in fact, your soapbox, and by definition you get to rant at will. I'm also guessing that by the intensity of this particular rant, I may be missing the more important issue that you might have a direct, personal reason to be ranting on this particular topic. If that's the case, I doubt seriously that you really want folks pointing out what they consider weaknesses in your argument/rant/choice of words.
Admin
Great point by Grubsnik, the generalists that actually make stuff happen are first to go.
The article missed the primary reason why I've left firms in the past - takeovers. I hate it when a successful, profitable company is bought out by a larger concern to prevent true competition.
After every takeover a bunch of shiny new managers arrive eager to disrupt the status quo. All that ends up happening is design by committee, outsourcing of sysadmins, a huge increase in bureaucracy which results in a change from a small, agile, profitable competitor into an inefficient cost-center. I always leave after a takeover, the only question is how long?
Admin
Very very very good article.
No, it's a terrible article. It completely misses the point of what "up or out" is, and why it exists, and then draws incorrect conclusions wrt programming jobs.
What is it? Every year, you either go up (get paid more, get a better title, get more responsibility, etc) or are fired. After five or so years at the firm, you automatically get fired if you haven't advanced to a senior (partner) rank, even if you are a net gain for the firm at your current compensation level!
Why does this insane model exist (firing good/productive employees with 5 years of experience?) Because employees beyond that level (e.g w/ 7 years experience) could quit and take your clients with them - so you have to make them partner (riches!) or fire them before they make serious client relationships. This is why up or out only exists in fields with high individual skills/training: lawyers, investment banking, some doctors, management consulting, artistic agency, etc.
Extra credit: if specialized talent is so important, why form partnerships rather than go it alone and get all the benefits from your talent? Answer available on request.
In programming, with little client interaction, the up-or-out model is a complete red-herring.
Admin
The value apex curve can be modeled by
= LOG(t) - kt + n
Admin
It was a fascinating article. However, the urge to quit is not that inevitable. Many intelligent and successful people in have stayed in their company for almost their entire career (Steve Balmer from Microsoft, Jeff Emult of GE etc.).
http://www.abhinav.sonaliagrawal.com/
Admin
It is exponential: x = 0 < y^? < 1. I usually just take 'exponential' to mean non-linear. You could also say, 'order of magnitude'
Admin
Maybe not directly related but .. still an example of sort of 'bad' HR management ..
At my previous Job I used to do some development and support/installations of the hardware and software on site. Which to me was ok, because i could at least do some development ... I always made it quite clear that I'd like to move more into development again. We got a new development manager after a while. He was ok most of the time even though we didn't agree on everything (particularly how great M$ products were ;)) - which in some way was good cuz it leaded to some interesting discussions ...
After the company was acquired by a larger one the development manager moved back to Pakistan ... and they then decided it would be cheaper to move all the development there as the salaries are lower there .. This was pretty much when i reached the apex as i'd been pretty much completely moved to support - which in the end i was only doing cuz i had to mostly ... Then we got some - rather crappy and buggy - CRM software which was .. well crap really ... So when it turned out i might not be able to renew my current tenancy and I was quite fed up with the UK in general already I didn't really hesitate to give in my months notice and move back to Austria. I offered them to help them out remotely for a while in regards to support and training,.. since 1 month was quite short to find a replacement and train him/her .. -- and there was only 1 other person who could do the support to start with ;)) ..
They more or less didn't answer / declined and didn't really start looking for a replacement until I was gone .. so in the end they were in a rather .. tight spot i guess cuz they had 1 support person who prolly wasn't able to go on holiday for like a few months ;) after a while they did contact me after all and we agreed that I'd help them out with some remote stuff and email support while the (still only) other support guy was on holiday .. after that I got a new job... I'm definitely not sorry that I left .. both the job and the UK in general ..
Admin
Well, a lot of people are nitpicking about absolutes here.
Lets review it again, with a pair of 80/20 googles on.
Personal development, and the dreaded apex 80% of the "unique" stuff a talented individual brings to a new company, is done on the way to the value apex. We are not talking about "grunt programming" here. We are talking about delivering knowledge of different ways of doing things, new processes, work habits, design patterns or just having the really good solution to problems the company is facing.
Once past the apex, your good ideas will become "old news" and be much harder to implement. Meaning you will at most accomplish another 20% difference.
Same goes for your personal learning. When you start at new place your learning curve is steep, the have new systems to learn, new ways of doing it, and a whole lot of other stuff. Once you've reached the apex, everything is routine. You've learned 80% of what you can learn from the job. The remaining 20% is what you get from staying there.
It's important to remember that regular training and education is not something that is centric to your current workplace. If you think the only reason you know c# or ruby or whatever, is because you got training on the job you are missing the fact that you could have learned it all by yourself or at any other job.
Staying or going 80% of the people who are leaving are highly talented people, 20% are "just the crowd".
80% of the people who are in their jobs for life, are the persons you would rather get rid of in a hurry. The remaining 20% are solid workers who keeps the company afloat.
Up or out The whole idea of doing a culture a of quitting, is manyfold.
The talented quitters [list] [] By making it acceptable to quit, the talented guys you want to keep for the longest possible time are actually more interested in staying on for a bit longer. Just finishing a major project or making sure they get a proper exit. [] Good exits are usually less costly than bad exits. [] People with fond memories of a former employee are generally beacons of recruitment. Friends looking for a job are generally pointed in your direction. [] Talented quitters who had a good time at your company and a nice exit, might be coming back for a second or third tour of duty. [*] Having an alumni network will also allow you to get "unique references" on applicants. This goes both ways (more below) [/list]
Overhead reduction [list] [] When people accept that they won't be the ones maintaining or doing the next round of development on a project, they actually take it serious when asked to write some proper documentation. [] People who leave your team on good terms, make much better trainers, when training their replacement [*] Focus will be more on getting people to become productive in a hurry, instead of doing the regular "expect people to be profitable after 6 months-2 years employment". [/list]
The Scrubs [list] [] If you have an alumni network spread out through the entire local industry, people won't just do a crappy job for 2 years and then quit, since they know it will follow them around for years to come, when applying for future jobs. [] People who know they will be let go in 2-3 years, won't try to make themselves in-expendable, or if they do, will be much easier to spot and stop, before it happens, which leads to the above point. [*] This also helps to make sure you keep the "ballast" to a minimum. [/list]
The keepers "Going up", doesn't mean, "going into management". If you spend 2 years just being a "developer", you can go on to do 3 years as "Senior developer" or "Architect" or "Team lead" which in turn lets you go to another job, which more responsibility. Not necessarily over other people. Keep this in mind when discussing "up or out". If you've spent 7 years being "developer", chances are that both you and the company would be better off with a change of pace.
Remember this whole treatise deals with the problem of high turnover, if you are not facing those problems (and don't have a huge dead sea, preventing you from hiring anyone new). Then this "solution" doesn't really fit your lack of a problem.
Addendum (2008-05-02 04:35): Bah, Thought they used the full suite of BBCode on TDWTF
Admin
Another way to look at this, is to think of "value" not as value to the company, but to oneself.
I'd like to know if anyone, using that perspective, find anything wrong about it. Of course, upholding the part that "up" does not mean changing careers into management or something like that.
Admin
I would just like to point out as a senior architect, I would never hire Alex or anyone who thinks this is a great or insightful article.
It's complete crap. A persons' value to a company only peaks if they let it, by becoming complacent or bitter.
Life is a learning process. There's more to be learned than can ever be learned on this planet, and there is absolutely no reason an employee cannot still contribute new ideas for the length of his tenure.
I've seen the "hotshot" programmers that most of you seem to think you're a part of. They come in, think they're gods greatest gift to mankind. They fuck things up beyond belief because they refuse to admit that they could be wrong. They cause more work than they accomplish, and then they skate off because they can spout off the latest buzzwords that impressed the next company down the line, leaving a wake of disaster.
At my previous company, they adopted a culture like this article endorses after I had been there for 3 years. It turned into a revolving door, average tenure for new employees was less than a year. Once most of the experienced developers left out of frustration, it was a complete disaster. No one knew what was going on, there was no architectural direction, and the quality of the product fell through the basement. They lost over 60% of their customers, and 80% of their sales. They fired the remaining experienced developers due to their high salaries, and the company went from 80 employees when I started, down to 30 today. They dropped all new development and are struggling to keep up with just maintaining the mess left by these so-called rockstar developers. They're not going to survive.
It's a shame too, because before they started hiring rockstars, it was a really great place to work at. We had a great culture, great atmosphere, great customers, and a great product. Now it's a complete mess.
When I hire people, I hire people who are looking to stay at their job and make the company a better place. These are the people who put pride and quality into their job. The rockstars don't give a crap, and will ALWAYS leave a disaster when they leave. No thanks.
My current company has been in business for 35 years, turning a solid profit every year with this attitude.
Admin
I have worked for 3 different companies in the past 5 years. I loved all of my jobs for the most part. During the annual reviews they all say how much they loved me. Then they say their standard raise this year is some crappy 3% and then I look at the market for my skill level with one more year under by belt and the raise should be more like 30%. (The first year it was over 100%.) What do you do then?
I have told them that it wasn't enough during the review. First company, they gave me a bigger raise (30% instead of 10%) on the spot but still not market value. Second time, at a different job, I was offered 3% after a great review and working rock star hours. I was pissed and said during the review I was going to start looking. They came back with a raise after about a month but by that time I already had half a dozen interviews set and a couple of offers on the table. I am not going to work for a company that doesn't know the value of my skill.
I agree with Alex that the good guys leave early, but probably only at cheap companies. I am making decent wages now, working half as much, and taking vacations now! I'm not going to leave unless I see another company that's going to offer me something better.
Admin
This is a very insightful article, it really has awoken some thoughts.
Admin
Grubsnik: I'm curious where the 20% figure comes from. I don't doubt it, but I'd love to beat management at my previous employer over the head with it...
Admin
It helps testers, technical writers, and new hires to understand the system.
It helps other programmers and tech groups build on the system.
It may even help in law, if you're suing the idiot bus driver who ran over your prize programmer for consequential damages.
Funny how five years ago we were all screaming about outsourcing, and now we're back to that old warhorse of industry turnover, isn't it?