4 ms·

Ask HN: Which technology will me reach independence?

7 points by crucini 15y ago

I've been doing unix server-side programming in the Valley for about 10 years now. I'm currently coding mostly in python but I've used most of the well-known languages (excluding Microsoft's creations). In the mid 90s I ran my own business for a year - not software related. It was stressful scary and fun. I want to start my own business. Not a startup with a one in a million chance of taking over the world, but a solid company making great products (or services) with a good chance of achieving 5 million in revenue. I am capable of working insanely hard (and loving it) when connected to a customer that actually cares about the product. Which has rarely been the case in the last ten years. What technical skills will help me reach this goal? Becoming a better and better server-side programmer seems pointless, since I already have the back-end toolset for any startup I can envision. To reach users, I need a good GUI. And all my ventures into GUI learning have fizzled out. To learn a major new technology I need total committment. I need to really believe that this is the path forward, or I will end up blowing it off. Should I learn Android or Iphone or modern web development? I've built web sites, of course, but I have not kept up with the modern javascript-based web. Which actually kind of disgusts me because it tends to be slow and annoying. I've dabbled in Android and didn't like it much (don't like Java) - I hear Iphone is much nicer to develop for, but of course it has restrictive policies and a declining market share. Or maybe I should bite the bullet and learn C#/.NET. I have managed to stay away from Microsoft stuff almost completely - I strongly dislike it - but it still seems to be the norm for niche business apps to be Windows only. It can at least provide a nice performant native-looking app to the vast majority of business users. It has dawned on me that an expert Windows programmer is better poised for independence than an expert Unix programmer, because he can deliver finished solutions to ordinary small businesses. Or even MacOS Cocoa programming. Or I could stick with what I'm good at and develop a prototype GUI in plain jane HTML, with some RPC to a controller layer; then hire outside talent to write the web 2.0 or Iphone GUI. One problem here is finding trustworthy talent; after 10 years in the industry I am pretty suspicious of programmers' abilities. I should also mention that any half-baked cross-platform GUI system is not going to appeal to me. I've wasted a lot of time on these things, only to hit dead ends. If I'm going to produce native apps, I want them to look at least as good as the competition. And I want a huge dev community so I am not the first one running into a certain issue. So which is the best step towards independence?

lsc 15y ago
>To reach users, I need a good GUI. I scoff. http://prgmr.com/xen http://prgmr.com/xen bad design is my brand. Now, I'm making rather less than five million in revenue; but I think I have a reasonable chance; I'll eat my hat if Linode isn't twice that. (I'm not saying I'm as good as linode yet, just that I have a chance.) I mean, my horrible UI in billing /is/ holding me back, but I think the places where I've gone ahead and used a command-line interface where some of the competition has used pretty GUIs? I think some of that has helped me. I mean, like everything else, it depends on your target market, but me? I think it doesn't matter if your brand is ugly, as long as it is consistent. You need to know two things. Your product, and your target market. Everything else is peripheral. It's obvious that you need to know the product, but I think it's also key to find a target market that you can relate to. When I go to silicon valley functions, social or professional, I wear my 'prgmr.com' T shirt (and yeah, it's about as ugly as the website) and it's fairly rare that I don't run into someone that tells me they are a customer. Part of this is that I use my network, but part of this is that I'm selling to "my people" - the sort of people I'd meet at professional or social events anyhow. I mean, almost none of my sales are done in-person, but I know the culture of my customer very well, and this makes marketing immensely easier.
- crucini 15y ago
  The site doesn't seem like bad design to me, just low-key design. It's obviously aimed at a technical audience. It also communicates "low overhead". I found it pretty appealing. What really makes the site feel credible is the candid mentions of high margins on VPS, higher bang for buck with colocation, and specific facilities in the bay area.
  - lsc 15y ago
    Thanks. I think that sometimes I go overboard with the notes I have above and below the pricing box; but yeah, clear and minimal was my intent. (well, except for the figlet logo. That's harder to read, but eh, it does really appeal to me, and I think it makes a unique and memorable logo.)
- twog 15y ago

lsc 15y ago
>Do you honestly think that a good design wouldnt seriously increase revenue? I do honestly think that consistent branding increases my revenue. Does it increase it more than bad design decreases it? I don't know. Better design would certainly increase conversions; problem is, I'm getting signups about as fast as I can handle as-is, with the current (poor) state of my automation. Getting more mainstream customers with mainstream expectations would also dramatically increase my support costs. There are two issues here. 1. my website is ugly, and 2. my billing and web selfservice system is in many places outright broken. 2. is a pretty big problem, I agree. (to the extent that it is difficult to read, that goes in this category, too; I honestly think I'm laying out the pricing information in the most clear way possible, though I could be wrong.) Now, website or no, my biggest problem right now is that I have really shitty automation, and my shitty billing system is part, but not all of that. I'm seriously considering going to a command-line ssh-accessible billing system; my ssh-accessible self-service systems all seem to work pretty well. either way, the big problem that I have now is that it's broken, not that it's ugly. 1. on the other hand? I don't think ugly is a problem. Sure, most people prefer salespeople in nice suits; but especially in my market? if someone comes up to you in a suit, the assumption is, at best, that he's a business guy that doesn't know much technically. At worst, the assumption is that he's a "professional" and you will be lucky to come out of negotiations with a shirt on your back. Looking like a sales person is about the worst thing you can do if you are trying to sell to nerds. How much does this carry over to web sites? I think it does carry over some. I know that my friends would often scoff at sites that looked too "web 2.0" with the rounded edges and gradients when that was the new thing; and those pop-up menus are generally considered a sign of extreme bad taste. I mean, this isn't to say my site couldn't be improved. I've got lots of add-on asterisks that are perhaps poorly-considered. Readability is important, no matter who you are selling to. Yes, if I was trying for the higher margin 'business class' customers? yeah, they'd expect some suits. They also expect a whole lot more hand-holding than the nerds do, too. For now? I'm fine leaving those people to the competition. Right now, some of the things I do, like requiring a ssh key up front? that is in part to chase these people away. The way I'm setup now, they won't be happy with me, and I won't be happy with them, so I'm generous with refunds. My point is that while broken automation (and I have a lot of that) is really, really bad, simply requiring more knowledge to use the product (at least if that is industry-standard knowledge, and not knowledge specific to your company) is in some ways good. It lowers my support costs, among other things, though it also lowers the potential number of customers I can have. Now, of course, like everything else, this depends on market. To be clear, most people on hacker news? not primarily nerds, culturally speaking, but most of the negative aspects I assign to my "business person" stereotype also don't entirely apply. Most people on hacker news are at least half business people, and as such value confidence and professionalism. Actually, that's part of what I like about this crowd (and the hacker dojo crowd) - it's sortof like dealing with business people, but they have one foot in my world. I'm slowly becoming more of a business person, and it's an interesting place to be, to learn, and to figure out how far down that path I want to travel. And I think it's very healthy for me; I've never had much respect for people that were culturally speaking business or sales focused; and around here? I meet some people that are what I'd call business or sales focused who are also technically better than I am. For me, the experience has been a little like when I started meeting women that were way technically better than I was; It's helping me break down my irrational and unproductive prejudices. I mean, as you can see from this message, those prejudices are still there, but I'm making progress.