1 |
Am 27.05.2012 08:22, schrieb Grant: |
2 |
>>> I'm debating whether I should hire an expert programmer for $X/hour, |
3 |
>>> or a company of expert programmers for $2X/hour. It makes sense from |
4 |
>>> a financial perspective to hire programmers directly, but I wonder if |
5 |
>>> there are benefits to hiring a really good company. |
6 |
>>> |
7 |
[...] |
8 |
>> |
9 |
>> For starters, you could give us a bit more insight into the kind of |
10 |
>> project we are talking about. What's the expected development effort, |
11 |
>> what are the services you pay for (binaries, source code, testing, |
12 |
>> maintenance, ...)? |
13 |
> |
14 |
> The project is made up of various and ongoing scripting tasks for a |
15 |
> relatively complex website. |
16 |
> |
17 |
>> Regarding programmer vs. company, I'd say it depends on what you expect |
18 |
>> and pay for. If you just want it coded, then the lone programmer is |
19 |
>> probably as good as the company (since programming itself doesn't really |
20 |
>> scale well with the number of devs). |
21 |
> |
22 |
> That's a really good point. |
23 |
> |
24 |
[...] |
25 |
> |
26 |
>> But in the end, these issues a minor. It really boils down to whom you |
27 |
>> trust more. Ask for references, look at their previous work, talk to |
28 |
>> them, etc. |
29 |
> |
30 |
> Can you tell me what sort of positive and negative things to watch out for? |
31 |
> |
32 |
|
33 |
I probably don't have enough experience to give you an exhaustive list. |
34 |
However, since this is a web development, the two biggest points I'd be |
35 |
looking at are: |
36 |
1. How do they plan to separate the production environment from testing |
37 |
and development? You don't want to crash your site just because the dev |
38 |
is too lazy to test his changes beforehand. |
39 |
2. Do they have a basic understanding about web security? What |
40 |
precautions do they take with regard to XSS, CSRF and the classic |
41 |
injections (HTTP header, SQL, Shell, etc.)? Do these words even ring a |
42 |
bell to them? |
43 |
|
44 |
Methodology is also a good indicator: Are they happy hackers with no |
45 |
real software engineering background, then they'll probably be good for |
46 |
smaller projects but will break down on large ones where you need the |
47 |
additional management. On the other hand, if they throw only buzzwords |
48 |
at you, I'd get suspicious. |
49 |
|
50 |
>> All things being equal, paying 1*x instead of 2*x gives you the chance |
51 |
>> to pay another 1*x to a second developer if things don't work out with |
52 |
>> the first one. ;-) |
53 |
> |
54 |
> Once I need more than one developer (which could come sooner rather |
55 |
> than later due to the availability of these guys) am I likely to |
56 |
> struggle managing them? I've read a bit about "Agile" software |
57 |
> development and I plan to read a lot more. Is that the way to go? |
58 |
> |
59 |
|
60 |
Two independent programmers working on the same project? I wouldn't do |
61 |
that unless they know each other and have experience working together. |
62 |
If you need to scale beyond the capabilities of your contractor, you |
63 |
should definitely start with a larger contractor (i.e. the company). |
64 |
|
65 |
I cannot give you any insight on agile development. First and foremost |
66 |
because I've never worked agile (well, unless you count rapid |
67 |
prototyping) but also because that's one of those buzzwords that can |
68 |
mean many different things to different people. |
69 |
|
70 |
> Would hiring a company make management a non-issue from my perspective? |
71 |
> |
72 |
|
73 |
Not completely but it's definitely better than managing two developers. |
74 |
You should still try to be in close contact with them. See if they |
75 |
understand your requirements, watch their progress, look at their |
76 |
intermediate results, plan the final acceptance testing with them and so on. |
77 |
|
78 |
Regards, |
79 |
Florian Philipp |