Nearly 10 months after telling my partner I wanted out, I have finally exited the electronics business, having negotiated enough for my half that I will have made approximately minimum wage. Frankly, I think the business is worth ap proximately -400, the cost of dissolving it, so I'm happy I got what I did. Of course the business could turn out to be quite successful under its remaining owner. I'm betting against it, but it is not yet a foregone conclusion. In any event, it was extremely unpleasant in many ways, but undeniably a <lj-cut text="a valuable learning experience">valuable l earning experience. One of the major things it suffered from was "This Time For Sure" syndr ome under Alex's guidance. Although each product line had failed to make a prof it, the next one, which we were ideally positioned to devel op due to our past experience, was going to make us a killing. After the third time wasn't the charm, I saw the pattern. He hasn't seen it yet; in my last business conversation with him, the day I traded him a stock certificate for a cert ified cheque1, he was still talking about his n ext big idea which was going to really make money this time - at least, it was as soon as he sunk some more capital into R&D so he could build it, of course. I made a series of bad investment decisions, based on Alex's assessment of the situation which I was careless about independently verifying. I'm not s ure why being less diligent in investing money into a tiny company I owned part of than I would have been in buying stocks seemed like a go od idea. It was a sufficiently painful lesson, at least, that I doubt I shall h ave to learn it again, at least not in exactly the same form. Another thing I learned is personally taking small-value orders and shi pping products directly to customers is pure hell. I do not intend to start ano ther business on the basis of personally shipping small orders. If we can't rai se enough capital to hire someone else to do that immediately, the business is a no-go. If I'm doing it myself, my cutoff is $1,000 or more per transaction, an d preferably to "ship" software instead of tangible goods. Having done both, I assure you that the latter involves far less suffering. It's important to be respected by one's associates. I don't know where this went wrong; I think I was respected at the beginning, but wasn't anymore b y the end, leading to a rather hostile negotiation in which I was not perceived as credible and therefore had to follow through on several threats in order to b e taken seriously. This remains an enigma; I can point to a time when there was respect, a couple years ago, and a time when there wasn't, but I can't tell you a story about the process of going from the one to the other. I'd like to unde rstand this better, but acquiring more information about the past, when one was already there in person as a participant and still doesn't understand, is unlike ly. Debugging failed circuit boards actually can be fun, provided one is no t doing it in the middle of the night, on a tight schedule, with a headache from the flux fumes, and a stiff neck from wearing the respirator in an attempt to g et rid of the headache. One also needs to have adequate information about what the boards are supposed to be doing (for example, by having designed them onesel f). For the most part, however, this activity pays poorly unless one is the original designer, so in a way it works out. In a "repair by : board replacement" world, it is a refreshing change to be able to make repairs a t a component level. All in all, working in electronics has been less enjoyable for me than working in pure software. Although it had the satisfying aspect of being able t o deliver hardware, this was more than compensated for by all the headaches of h aving to deliver hardware. In the future I shall most likely focus elsewhere fo r my primary source of revenue, and do the occasional electronics project as a h obby, or the occasional firmware consulting job if there is a market for it. It's useful to keep track of how much time you spend by writing somethi ng down at least once per day. Subsequent recollection is susceptible to system atic over or underestimation, and you want to have the information about how muc h time you spent on the last project when you're trying to figure out how much y ou need to charge for the next project. A lesson which I have applied already to my next venture is that it is important to agree in advance upon benchmarks for the minimum achievement level required to continue operating the business, so as to make it less likely to get into a situation where some partners want to continue and others do not. Realistically, I do not think that having learned these things will act ually make any of my subsequent endeavors more likely to succeed. I do think, h owever, that it will reduce the cost of failure, and that in turn will make me m ore likely to eventually succeed by conserving resources and permitting for more tries. [1] The Firefox spell checker doesn't accept this spelling of "cheque". Perhaps next it will deny that "doughnut" is a word. Oh my. I thought I was joking, but it just refused "doughnut" in the previous sentence as well! Perhap s in the next revision they can continue the newspeakification by accepting "lol " and "<3" as words. To make matters worse, the configuration option which i s supposed to disable this reprehensible excuse for a "spelling checker" does no t work.