5 ms·
That equation is a theoretical one. On the whole, jobs that offer high $/hr will mandate crazy hours. Jobs that will allow you to go 20 hr/wk, or 10 hr/wk wil
by stakhanov 7y ago
That equation is a theoretical one. On the whole, jobs that offer high $/hr will mandate crazy hours. Jobs that will allow you to go 20 hr/wk, or 10 hr/wk will usually not pay you well on a per-hour basis either. -- I predict that in response to what I just wrote, there will be people showing off about how they managed to make a consulting lifestyle business like that actually work because of how awesome they are. -- But keep in mind they are the exception, not the rule.
- Retric 7y agoTaking time off between jobs is a real option for most programmers. Depending on how long you stay at each company it can even add up to a few years of time off without having obvious gaps on the resume.
- stakhanov 7y agoYou have to have the stomach for that... Personally, I couldn't get any real enjoyment out of a time period where I didn't know when/where/whether I have another job lined up, and opportunities to get truly trustworthy commitments on start dates that are a significant period of time in the future are rare.
- dmoy 7y agoI agree with you. Unless you had like, a lot of money.
- badpun 7y agoI do that all the time - like literally my CV has probably 5 or 6 gaps of varying duration (between 3 months and two years). Often, during those gaps I spend some time working on something tech related, so I put that into the gap in the CV. Even with those gaps, I'm having zero problem finding well-paying jobs. I think it's because, despite those gaps, I represent good value to the employer - I know the latest tech, have done good work in the past, can present coherent and mature opinions during the interview etc. Also, I only do long-term contracts where they explicitly don't care that I'm a job-hopping mercenary that is not ever going to believe the "company mission", because the assumption is that the whole arrangement is temporary anyway. In practice, they always want to retain me and I end up quitting out of boredom after a year of two. To summarize, the gaps are probably important only when you're in close competition with lots of folks similar to you. But, if you're a bit of an expert in a specific niche (mine is Big Data - Hadoop and related crap, applied in enterprise IT context) and there aren't that many competent candidates around, the gaps suddenly don't matter.
- rabidrat 7y agoAre you over 40 yet? It changed for me this last job hunt.
- Aeolun 7y agoIf you are still hopping by 40 they probably don’t get what they want any more, which is an experienced, stable worker.
- badpun 7y agoI'm 38. We'll see what happens in a couple of years. In any case, these enterprise companies usually have a position of architect on their teams, where grey hair are even an advantage. I don't like this job, as it's more headaches than coding in a lot of ways, but when if ages turns out to be a thing, then I guess I'll be architecting (I've done it several times already, so I have the CV for that).
- dkersten 7y agoI’m similar. I have one multi year gap and a number of smaller gaps. Nobody has ever commented on or asked me about them. I do put any relevant projects I’ve done in that time on my CV, if I have any, but I done’t always have much technical stuff to show for the time. I have started startups though and run a programming meet-up, in mostly asked about those.
- war1025 7y agoIf not for the fact that I need to have health insurance for my family and the only way to do that without breaking the bank is through an employer, I'd agree with you.
- alchemism 7y agoSpeaking of signaling mechanisms, if you take time off from working to travel to “interesting” places, and mention it in the interview, you are indicating a clear lack of financial pressure, which I have found to be quite helpful in price negotiations. You aren’t jobless; you are successful. Even as a bluff, with a now-empty bank account.
- elindbe2 7y agoTrue, but I'm looking at it more over the course of a lifetime. If I need $X to cover all expenses in my lifetime then I need to work $X/Y hours where Y is my earnings per hour. This is a bit oversimplified if you include savings and investment but in general is how I look at it.
- lotsofpulp 7y agoThat "$X to cover all expenses in lifetime" figure is impossible to calculate, and is not under your control at all. It will be inflated away without your consent, and requires constant evaluation and re-evaluation. My philosophy is to always be investing in capital, whether it be equity, political (relationships with people), or resources.
- elindbe2 7y agoMy spending is not impossible to estimate and for the most part it is very much under my control. If what you said is true then no one could ever reasonably retire at any age.
- lotsofpulp 7y agoThe relative security afforded by various factors in developed countries (especially USA) since after WW2 up to now may not be available going forward. The biggest challenge is the increasing proportion of non-working people to working people, at least partly due to aging and fewer children. The part of spending you can't control and which happens to be the largest expense is medical care, which is not a problem right now due to Medicare, but it's already stressing the system and we have quite a ways to go before the aging population maxes out. Retirement ages are already going up and up, and I can definitely foresee rationing of medical care as we push the system to its limits.
- Karrot_Kream 7y agoThe closer you get to retirement, the more obvious it becomes about how much to save to reach the goal. Until then, it's better to just try to eke out as much savings as possible than estimate your spending. Estimating lifelong spending is quite similar to modelling global economic activity over a long period of time.
- wolco 7y agoI find 10 hours at high cost easier to sell than 80 hours at a high cost most want to pay for 25 of thoses hours.
- stakhanov 7y agoI agree. There's probably a dead zone somewhere in the middle. One side of the deadzone there's an equilibrium doing fulltime at a mediocre hourly rate. The other side of the deadzone there's an equilbrium doing a very small number of hours at an exorbitant rate. But for that latter model, I would expect a significant number of non-billable hours for project acquisition. (Maintaining client relationships outside of paying projects, doing sales work around new leads, doing stuff like PR and writing articles and whatever else you do to be recognized as an expert/authority in some field, etc. etc.) so that the billable and non-billable hours probably also average out towards a mediocre effective hourly-rate. Early in your career, that probably means you have to do crazy hours there too. Later on, you might get dividends in the form of not having to work so hard on sales any more.