Mothers should be programmers
lepid0ptera.livejournal.com
lepid0ptera.livejournal.com
The first is distraction. When I am working on a difficult problem, it's true that I often need periods of inane distraction interspersed with periods of extreme focus. In theory, taking care of a child and hacking could be combined; where I go for walks or play video games now, I could be playing patty-cake. But those breaks need to occur when I intellectually need them, not when the baby needs them. When I am madly coding something I understand well, I very much need to avoid distraction. When I am mulling over something I don't understand, I very much need to embrace it. So I suspect this would not work well in practice.
A second problem arises from simple physical compromise. I am presently seven months pregnant, and--much as the admission grates on me--I have found my performance slipping throughout. Sometimes because I am tired, sometimes because I am forgetful, sometimes because I am more excited about the coming baby than the current project, sometimes because half a dozen random things are wrong with me physically and discomfort is . . . distracting. I had hoped to keep working at full capacity, and while I've kept working full time, I have to be honest: I haven't been as effective as I once was. Where I used to produce a couple brilliant ideas a day, now it's more like one or two a week. I don't see that improving just after the trauma of birth, during the subsequent hormonal adjustment, and what with the excitement and distraction of having a baby around.
Programming is an activity that depends so much on mental sharpness; where caffeine and sugar are performance-enhancing drugs, insomnia and strange hormones are remarkably destructive. On a bad day, I can spend 8 hours not-quite-achieving what should be half an hour's work. Mundane programming I can usually still do, but great hacks? Not so often anymore. Nor do I see them recurring with regularity for a while. No doubt about it, being a mother is a sacrifice intellectually. All indications are that it's well worth it, and I wouldn't do otherwise. But it strikes me as unrealistic to suggest that it's a light side-activity that one can casually combine with a serious commitment to work. Even very intellectual work.
I suspect that programming and other forms of concentrated work will require a phase shift in order to continue a high level of productivity, and the type of productivity won't be the same.
But, as DHH of 37signals points out, people don't typically use their hours well anyway. If raising a child left only two hours a day to work, that's still 10 hours during the week -- as much time as DHH spent coding basecamp and rails his first year. If you add to that time you can spend planning, communicating, and thinking about the problem (to fill the time when you're more likely to be interrupted) you can optimize those few hours even further.
So far as I can tell, this is the pattern my mother used while she was caring for me and working her way through journalism school. I had colic. But she'd do her interviews in the intervening time, and she'd collect all the information she needed, and wrote fragments she wanted to include. When she got the chance, she sat down and wrote the whole article in a single sitting.
PS: Thanks for being part of this community. I started to read your comment page, and you've contributed many wonderful insights I've had to copy to remember for later :-)
I follow a similar pattern at work these days. Having amassed a large amount of technical knowledge, I can spend time I can't spend innovating in teaching, documenting, cleaning up code, running test cases, doing research, doing arcane analysis. And while one or two brilliant ideas a week isn't what I'm used to, it's not nothin' either.
But oh, for those two-day coding runs of yore, during which it seemed I could move the earth . . .
A good manager is going to cut a late-term-pregnant developer a lot of slack, because "below potential" is not the same as "net negative value", and because proven developer talent is very hard to acquire and even more expensive to ramp up.
You worry about performance changes that are long term, systemic, and corrosive to the project. You worry less about easily explainable short term problems that you have 2-3 fiscal quarters to plan for and that run on a set time frame.
Programming doesn't suit people who are constantly distracted; the way I cope with others' presence around me is to completely ignore them. I can have conversations for hours on end while I program and not remember any of it. I agree to things, laugh with jokes and listen attentively. All of it a facade, of course; I don't really care about others when hacking.
Days run into each other sometimes and you have no idea what went in the world. I wouldn't do this when I have a baby, and I don't recommend it to new parents. Do marketing for 2-3 years, instead, and give that kid full attention when you go home.
I mean, sure I agree, in certain aspects of the job there is a level of concentration and 'state' that you have to maintain in your skull in order to be able to program properly.
But a certain amount of the required 'state' can be alleviated by having a pad of paper next to your desk and writing everything down. The rest of it can be dealt with by writing code that has a simple and sensible structure.
And honestly, the majority of my coding is mind-work. I don't have to sit at a computer for 5 hours to be thinking about that piece of code I'm going to write. One I have read the right documentation and I have figured out what the problem is, I normally need a bit of time for it to gel, then I type it into the machine.
I think the point of the post was that new mothers need time to bond with their infant. 'Doing marketing' for 2-3 years instead implies to me that they are working the 9-5 grind (which a programmer can theoretically avoid). Which misses the point entirely...
There's no reason you can't sit at the computer with the baby right next to you.
My workday normally starts around 7:45 AM. On a good day, I leave for home around 7 PM. Sometimes I get to leave "early" by 6 PM. I usually break for about 1 hour (or less) at lunch time.
I do not wear T-shirts or tennis shoes to work. There is one other programmer who does; his contract has not been extended past the end of this December. I do work from home, but only because I don't get all of my work done between 7:45 AM - 7 PM when I'm at the office.
I could not use an OLPC formatted with Breezy for my work. I use a 2.26GHz Core 2 Duo w/ 4GB of RAM, and I still find my computer is too slow and doesn't have enough RAM to, for example, run more than a few VMs at once.
I do often work from 2 AM - 4 AM at home when inspiration strikes and I cannot sleep. One reason for this is the lack of distractions at this time of night: my wife is in bed sleeping and there are no coworkers to stop by and try to chat.
However, I could not work from home and satisfy my job duties. There are too many times when I need to work in person with, for example, a professor having trouble with one of the programs I've written, or go to a meeting with the deans of some of the schools at my university.
I also don't have sufficient bandwidth at home to work satisfactorily. I often do tasks where I need a 100Mbit or 1000Mbit pipe to my workstation; a 10Mbit connection is not sufficient. (For this same reason, I don't sit outside on the commons and work on my laptop wirelessly.)
Because of these reasons, I have chosen not to have children at this point in time, as I don't feel I could adequately give them the time and attention they deserve. It would be really nice to take off a few months of time to spend with my wife & my family, and have my job guaranteed when I choose to return, but that is simply not an option I have.
(For the geeky amongst you, the thing I do that needs bandwidth is mostly using TNS for running PL/SQL; running TNS over a 10Mbit WAN is frustrating at best, and completely useless at worst. It is not suited towards dealing with terabyte-sized data sets.)
Also, what you are describing is a job in a bureaucracy. This kind of environment isn't something do be proud of, it's something you work despite (though this is my personal preference, and there may be people who like this).
The point of the article isn't that it's easy for any workplace to accomodate a mother working from home. It's that it is doable. This depends on someone higher up in management having the will to do this. But this is a management problem, not a problem of discrimination against women.
Basically, this is not a problem if the people higher up in the food chain would start treating their subordinates as human beings. Which, sadly, might just be too much to ask of a lot of managers.
As far as the bureaucracy goes, that is one of the pitfalls of working as a staff programmer in academia. The point of my post, however, is that many staff IT or programmer jobs are simply not compatible with the kind of working environment lepid0ptera describes, and I think this extends outside of IT or ERP style teams at large instutitions: my friends in the game or design industry frequently pull long shifts when trying to meet hard deadlines. This kind of environment is what is hostile to parenthood in general (not just mothers trying to take maternity leave).
So I agree with you, and I'll up that to say that things don't really get easier as they get older, because they sleep less. You'd think the older they get, the more time you'd have, but, at least at 4 and 7, it really just means they can climb higher, fall further, hit harder, and break bigger things.
But one main problem is when children are young mothers are much more likely to be tired or worn out running on little sleep and probably have less time/will to work on complicated problems
Believe me, I've tried doing programming work during periods while watching our young son, and it is nigh impossible to do any heavy duty work unless he is sleeping. Writing one-line unit tests is doable. Debugging complicated code is just barely possible if you are lucky. Breaking new ground is downright impossible.
I suspect the author of the original post is not actually a mother...
When you have less time to hack, you tend to use it more efficiently (I've worked at home most of the time my youngest 2 were babies).
I can't imagine having had a nanny and missed out on all that bonding, fun, and awesomeness that is baby/toddlers.
(Not to mention things like my son puking up a full bottle of milk straight onto the keyboard of my laptop - I was feeding him and coding at the same time, which was probably a bad idea).
Of course, no nanny costs more than the most junior developer makes.
I'm not sure how toddler awesomeness is relevant to the discussion. Nobody's offering a free lunch. But there clearly are jackasses who don't want to hire women because they think maternity leave is going to screw them.
Insurers are freely able to discriminate, it seems like employers should be able to as well (I know in this age of political correctness it's impossible to).
Should terminally ill people be able to go on apprenticeships? I'd say on balance no - if they die before they're trained in the job, then the employer who is footing the bill of training them, has lost out.
Or at least, that's been the case with me, and with every new father I've ever worked with or known.
In a two-career household, you are expected to figure out how to allow both parents their careers. If the mom is the only one awake at night, you're doing it wrong.
New book from O'Reilly: Hacking for Housewives.
So what? Both men and women bond with their infants. One, the other, or both will inevitably sacrifice bonding for income. Women with full time jobs who want their kids breastfed, pump.
Why do women need special arrangements to rejoin the workforce? Doesn't that simply perpetuate the gender discrimination we already have in the industry?
It is unlawful to discriminate against female candidates because they might take maternity leave. It's also stupid: whether they take formal leave or not, the performance of a new father is going to plummet as well. Let's leave it at that.
The fact is though that most mothers aren't interested in programming :/ Something that doesn't really need 'fixing' (Because it's up to them what their interests are).
I was speaking to a female VC a few weeks ago on being a professional and having kids. She told me flat out not to think about having kids until my thirties (if ever), because it's the only way. And even then, it's going to take a crap load of time away from your job.
Here's a news flash: if you're an entrepreneur, your life will be no more or less accomodating to children when you're 35 than it is now.
Diversity. A diverse workforce is a strength for a company. Excluding moms would cause a weakness in the corporate culture which may go entirely unrecognized, even if it ultimately helps contribute to the failure of the company.
One thing this particular form of diversity could help address: The fact that females play games a lot less than males as well as generally have less interest in a variety of tech products. If there were more women and mothers in programming, the odds are good that apps would be designed in a way which appealed more to the female market than many of them currently do.
Company policies that are friendly to moms and their kids tend to also be better climates for anyone who can't make it on "traditional" terms of success. Finding an effective means to accommodate the rather large issue of motherhood would tend to automatically create a company culture and set of policies that resolved many lesser issues with little or no additional effort. Happy workers who don't hate some aspect of their job are generally going to do a better job. If the working climate is vastly superior to that of your competitors, you will likely also garner high levels of employee loyalty and a lower turnover rate. Lower turnover pays dividends in many ways, well beyond just the obvious savings of not having to spend so much time and money on the hiring process.