guyinpv 28 Light Poster

Just as an update, I am expirementing with Dwolla. They are an ACH direct bank transfer service and the entire fee structure is 25 cents to receive anything over $10. Any amount under $10 is free. That's it!
The only hard part is convincing people to sign up and pay this way. It is totally secure in the sense that nobody has to give me their credit card over the phone, or use my sketchy website or what have you. In other words, it works really well for everybody signed up to the service. If you go to "send" money to someone and begin typing their email, it automatically starts finding matching people in the system to send to.
I think it has a lot of promise, and connecting a service to your bank shouldn't be that scary, since people use many services that do this anyway, like Mint or Quickbooks or any other finance tool. Even Paypal and other services connect to your bank for transfers, so there is nothing all that scary about it really.

I also use Nutcache for some business stuff, and they just recently announced integration with Stripe! So now with Nutcache and stripe I can still send out invoices and get paid with credit cards (with blasted fees), but if my clients choose Dwolla, I pay just 25 cents to receive any amount.

Those are two options I thought I'd share.

guyinpv 28 Light Poster

Hi, I'll try to make this quick. I'm used to programming PHP in context of building websites, to add random bits of dynamic functionality, but every once in a while I need to build something like a complete application. My problem is this, as a single developer, there comes a point in time where my project starts to get too big for me to manage in my head. There is too much logic flow, too many data points to track, too many twists and turns in the flowchart. I start to get lost in how many branches of logic the program has taken. For example think something like "ok the user did this, they have that permission, they can go here, but only if they've done that, they can't see this thing yet until they click here, but they need to enter this data first, but check if this date expired and they are still valid..........." Just making that up, but you get the point, I go so far down the rabbit hole of logic flow that I can't remember where I'm at sometimes! Then I have to go backward and figure what to do if they made some other choice and go down a different path.

The point is, I don't know how to manage all this logic, to keep track of every possible action and reaction of the app and all the hundreds of conditions that might be going on. Worse still, if I leave a project for …

guyinpv 28 Light Poster

My general method is something like this:

step 1: Open session and see if a user is active.
step 2: Perform your routing functions and see if the user has access to the page requested.
step 3: Allow access, or show a "not allowed" message, or even simply show a blank white page with no information at all, or redirect them, if not allowed.
step 4: If no user is in session already, then either show the "public" information, OR send them instantly to the login and register page.
step 5: process the login, test if user exists, etc etc.
step 6: Return fail if bad login. Here you could track IP addres, log how many login attempts were tried, perform login timeouts etc.
step 7: If user matches, put the necessary user data into session and redirect them to the start page of your app, they are now logged in.
step 8: If they click a logout action, get rid of all that session/cache/cookie info to make sure they are logged out, and then redirect to public page or back to login screen. Search online for best way to destroy session.

For the incidentals... You need to be very careful how you return failed login attempts, never show things like sql errors being returned. Don't hint whether they typed a username right but pass wrong, or vice versa. Don't have one error that says "no such user" and another that says "wrong …

iamthwee commented: nice +14
guyinpv 28 Light Poster

Hi guys, I'm looking for comments and suggestions regarding the use and place of jquery/ajax in site development.
I know jquery allows a lot of cool stuff and so does ajax, but when a site becomes heavily based on these tools, the use of PHP dwindles. I am used to PHP being the engine of a website but in order to use Ajax, a lot of site issues are handled in javascript instead.
I can inject php variables into javascript code because PHP is interpreted first, but at the same time, any new dynamic content that comes from javascript, PHP no longer controls or can handle, at least not without a page refresh.

When I use ajax to get dynamic content, I can't create URLs to return to these screens without a lot of finesse. When I use PHP for "dynamic" content, I can't really use javascript, because jscript doesn't do it's thing until after PHP is interpreted.

The problem is I'm starting to get frustrated trying to use the "coolness" of javascript but maintain a solid PHP core. In PHP where I would normally include another file to run as-is, to use ajax I have to include files that output html or json data to parse. So I find myself having some files here for PHP-only processing and some files there for PHP/JSON/AJAX processing. I have elements on the page that are setup with PHP code and then other elements that are managed by jquery dynamics. To …

diafol commented: I get you +14
guyinpv 28 Light Poster

I went through all the extensions folder removing addons and extensions to see if it would work, it did not. I checked for read only and other permissions on folder structure, no good. I created a new user account and ran it in that, still no good.

I'm running out of ideas! I need this back!

guyinpv 28 Light Poster

Be careful hiring any SEO "consultants or companies", a lot are scams. You'll know by the promises, i.e. "top rank by Tuesday!" "Submitted to a 100 million search engines!" and so forth.

The first way is to make sure your site is designed right. No duplicate content, proper robots file, appropriate site layout, perhaps a sitemap that bots can find that is pointed to in your robots file. And though controversial these days, it doesn't hurt to drop some proper meta tags for site description and so forth.

Second, proper code, use alt tags, give links descriptive names that match the content they are linking to. Use as much text as possible rather than images, as the words "my cool stuff" in an image can't be seen by a bot, but real text can.

Third, don't cheat. Don't keyword spam your own site, put white letters on white background to try and "hide" keywords, you'll get punished. Review the criteria that Google and other search engines have available that tells you what not to do to get punished.

Fourth, drive traffic to your site, get your links on other sites more popular than your own if you can. Bloggers and the like. Search engines will like you better if higher ranking sites link to you.

If possible, one or more of your primary keywords should be in your domain name. If two people sell pizza, then "bobsgoodfood.com" will likely rank lower than "

cguan_77 commented: thanks very good +0
guyinpv 28 Light Poster

Browsers don't show scroll bars for background images, only the block level objects. If you use an <img> or a table it might show scroll bars. But using background image it shouldn't.
If you put the background on <body> you should be fine. If you put the background in a <div>, you'll have to set a height/width on the div or it'll shrink to nothing because background images don't stretch their containers.

No harm in trying it, pretty simple to test! Good luck