Ask HN: How Would you Implement this? | Hacker News Reader

Ask HN: How Would you Implement this?

1 pointlouisadekoya5 comments

I am not a developer but I have dabbled in the past and I consider myself fairly technical and web savvy. I have an interesting idea for a small web app and I am considering options for implementing it.

So I would like to ask the HN community how you o glorious hackers would advise me to implement such a lowly app as the one (sort of) described briefly below.

Features: 1. A main page that allows the user to upload as many images as they want. Kind of like selecting multiple files to attach to an email. 2. The user should be required to enter an email address. 3. The site will charge per image so needs checkout facilities including taking shipping details and postage charging etc. 4. At checkout the user should have the option to: 4.1. Save their shipping address for use next time (in other words sign-up). Otherwise, the user should not need to sign-up to use the service. If they do choose to sign-up, they will need to provide a password - ideally without going to a separate page. 4.2. Have their images stored for future use and shared with their Facebook friends. This would require d user to enter their Facebook username (email) or state that it is the same as the email address already entered above. 5. Ideally, all of the above functionality should be on a single page. 6. A My Images page for registered users. 7. A Tell a Friend page and associated form. 8. An About Us page.

Without giving the idea away, the closest thing I can liken this to is one of those websites to which you can upload your photos and have them printed and sent to you.

As a non-developer on a very tight budget, below are the implementation options that I can see for now: 1. Buy/clone one of said photo-printing services, modify it and for the Facebook functionality, integrate/mashup the site up with Facebook Connect, Gigya’s Socialize or WidgetBox’s AppAccelerator. 2. Use one of the many easy website builders out there like Weebly or Synthasite, mashed-up with form-builder Wufoo and do the same as above for the Facebook bits. 3. Pay someone to build the site from scratch and integrate with Facebook as above. 4. Pay someone to build it as a Facebook app directly.

So what do you think? Which of the options are viable/best, how much would you charge a customer for option 3 or 4 above?

Option #3/4: In a software project, you have three main levers - time, quality and cost. Improve one, then you need to lose on one or both of the others.

If you want real ballpark to start the thinking process (can't really promise much more than that without a lot more information) - well I'd say that a local 2-3 man shop might quote around $25k on this (i.e. ~4 weeks work).

Course, people will tell you that this is 1 week's work, or 2 days, or whatever, but generally this will be coming a developer, and they'll only quote the development time.

This cost will vary greatly depending on location (e.g. India vs Bay Area vs elsewhere). You could look at going somewhere like India, China or the former Eastern Block for development - the daily rate will be lower, but I suspect the overheads will swamp you, particularly for a one-off piece of work.

The approach will make a big difference too - I suspect you'd find if you tweaked certain requirements something like Google Checkout could provide a lot.

If you wanted to build something to demonstration level only, then probably half that.

A larger professional development shop could easily quote 50-100k. They'll want to do requirements, testing, etc, etc.

If I was building this in/for a corporate environment with all the trimmings, then I'd push this ballpack up to 100-200k. (You don't really get "Hello World" in this environment for less than 100k - but it'll be tested to the n'th degree, scalable, fault tolerant, etc, etc).

There are other costs as well, perhaps more significant and something to be aware of.

1. How much interface/web design do you want to put into the product? You can spend very little through to astronomical amounts on the design.

2. If you're handling images, potentially large ones, then you're going to have a much higher bandwidth and storage requirements than many web apps.

2B. If you need to scale (i.e. X new users every day) the you'll need to accommodate that. Even if you're not paying for the capacity, building in the ability to scale can be expensive. This could be anything from using a top-tier host, or investing a lot more in development and performance testing.

3. You might need other things like domains, secure certificates, company formation fees, legal fees (disclaimers, etc). You might (and I usually recommend) want a good copywriter. These things are usually pretty manageable in themselves, but add up really quickly.

Thanks for the thoughtful post. Wow - that much huh? It really feels like a fairly basic app to me but I guess that shows how much I know. I did have someone build me a comparably complex app for me for $600 last year though - but no facebook integration with that one.

Yes, it had occurred to me that storage could be an issue with this one.

By the way, do you know any good resources for finding good but affordable copywriters?

Still keen to read others' thoughts.

You're using one now. I'm in the midst of launching a site myself (outsourced as I'm in a situation much like yours) but I have 9 years worth of copywriting/marcom/biz dev background. Feel free to shoot me an email and I'd be happy to chat.

Well, the idea comes to my mind that, if it could be done for $600 (or even $6000), then it's a side project that any developer, anywhere could do in a month of spare time if they had the idea. You can't patent/trademark an idea either so, if you can build it for cheap, someone can copy it for cheap as well.

Looking at your specs, I don't necessarily agree that you'll need $25k minimum. It seems pretty simple to me too; $1k-$3k for the right person seems right to me to get you your beta then another $1k for bug squashing and $x/mth for maintenance.

What I don't get is that (a) you aren't the developer, (b) you aren't wanting to write copy, and (c) it doesn't seem like something that needs a salesperson making calls, so what's your role? Rhetorical question really.

Yeah, $25k may seem high, and probably varies a lot on location. $25k is what I would recommend if someone was in the early stages of a business case.

$600 is really only a day or two's work for a professional developer. Depending on where you're at, you could spend double that just nutting out the exact requirements.

That said, as my first line was trying to allude to - it's horses for courses. You can go a lot cheaper, or a lot more expensive depending on what you're after.

If you have the right contacts, and you want something whipped up, $1-2k might be more than enough.

Reply on news.ycombinator.com