Convert Software Net to Internet

[MARCO RUIZ VENTURA]

I have a software of billing that works in net. All the users are located in oneself building.

These users are reorganizing now in 3 different locations, all them to several kilometers of distance of the other one. The offices will connect for internet.

My question is: What is what I have to modify in the software of billing in net, so that it can continue working, now in internet?.

I don't have knowledge in programming in internet, for what I request their help.

Thank you.




Spanish :
Tengo una software de facturación que funciona en red. Todos los usuarios están ubicados en un mismo edificio.

Estos usuarios se están reorganizando ahora en 3 ubicaciones distintas, todas ellas a varios kilómetros de distancia de la otra. Las oficinas se van a conectar por internet.

Mi pregunta es : Qué es lo que tengo que modificar en el software de facturación en red, para que pueda seguir funcionando, ahora en internet ?.

No tengo conocimientos en programación en internet, por lo que solicito su ayuda.

Gracias.

[Paul Purvis]

Hi Marco,
there are about as many ways of doing this as there are to skinning a cat.
i am no expert but have had some experience using the internet.
you mentioned billing, how much billing goes on and how is your program built, i assume you wrote the billing program.
i will make some suggestions to you even though they might not be the best.
the first thing you need to consider is that a connection by internet may not always exist due to problems out of your control.

depending on volume of billing and how integrated(by what means does your data get built) your accounting(billing) system is makes all the difference.

if you wanted to link those systems, look into a VPN type solution over the internet while using quality equipment(do not with cheap equipment here).
you programs may not need much changing using this approach and you just might have to build into your software some kind of assurance all data over the internet gets transferred before making a posting(billing) and also a way to reprint a billing if your internet connection gets broken.

you could go wireless from one location to another, you would be shocked at the distances and speed you can do with wireless with the right equipment, that you are not going to be able to get over your basic internet connection and you have control over your own equipment, not some company that might be down, although you have to consider backup equipment when doing this kind of communication structure.

a better way to do billing, and it all depends on what you are doing, is to keep transactions off site and have a program to transfer that transaction over to a central data location for processing, you want to storm proof your system, so to speak. a good design upfront will last for years, no matter whether your offices are at the north and south pole.

[Paul Purvis]

i just purchased a apple computer, actually a Mac mini. If i read the correctly, there is a program call iDisk, it looks promising on some features i would like if i read it correctly.

if i did read it correctly, you can setup with the company Apple a location to keep files.
then place a file into a specified directory and the iDisk program will load the program to a location Apple has and it can be shared by others and also the files at Apple's location can be password protected.
even if a person did not want to use this as a main service, it could possibly be used as a way to share files as an alternative to some other service.

also Apple has some nice built in programs to do backups called Time Machine that would supplement a Mac machine being used as a server of some sort. I think the MAC mini was a cheap way of testing what is going on in the Mac world, so far i am very impressed. i did not want a laptop because i do not plan on using it as such and because i can actually load windows on the Mac mini and run it at native speed, i will not loose an investment even if i use it just to be a windows computer.

one thing i have already tried, if the mini losses power, you can set it to turn back on once power is restored. you can also set it to reboot at certain times, sleep at a certain time, boot up at a certain time, shutdown at a certain time. none of my laptops do that. there is also raid support on computers that have more than one hard drive. in case you did not know, MAC OS X Leopard is BSD certified now.


paul

[Joe Byrne]

Hi Marco,

Its logical to assume the physical layer (your actual Internet connection) is already in place, or will be, so I'll forgo any discussion on that matter.

The first thing to understand is that you're not going to be able to "just modify" your existing application to magically run over the Internet. There will be a fair amount of new coding, depending on how you've designed it already.

You have two main methods. You can rewrite the app to use a web browser as your interface (probably the biggest amount of recoding necessary), or design the existing app to access a central database. The main consideration is to minimize the amount of data that needs to cross your (slow) Internet connection. (remember, you're moving from 10mbs (or more) to 1.5mbps at best. Other than dedicated fiber, DSL, or a T1, you'll probably have a decent download speed and rotten upload. You'll need both for a "business" app).

Some databases have TCP/IP communications built in. This will make your life a lot easier, but you may have to rewrite large sections of code to use a different database. Tsunami (with Don Dickson's TTDS or the promised upgrade with TCP capabilities built in) is a good choice, as would xSQL or SQLite.

Another thing to consider is security. Whatever you send across the iNet can be, and likely will be, snooped on. You'll have to decide how you want to encrypt the data between the two points. You can encrypt the data locally, send it to the "server", and decode it locally, or you can set up an encrypted tunnel (a VPN) between the two sites and let the hardware (routers) do the encryption.

However, based on your limited information, it would be impossible to give you much more. You'd have to describe how you store and retrieve the data currently, how many transactions (on average) you'll have simultaneously, the size of the data packets (records) being sent, and so on.

Generally speaking though, I'd start getting mentally prepared for a fairly hefty re-write most likely.

[Paul Purvis]

also worth mentioning is remote control software and terminal services which i know hardly any thing about.

on the remote control software, i use pcanywhere.
this way, you would setup a windows computer on the central end and have the remote location computer operate the computers at the central office.
the only problem i have with pcanywhere is the screens do not have perfect characters on a console based program but it does great running a gui/graphic program on the central(host) side. the remote location is called a client.
depending on the screen size you have your host computer set to, makes the biggest difference in how fast the screen updates are on the remote, but it will be fast enough for user input. the application is running on the host side, so if a disconnect happens, all the user at the client side has to do is reconnect and they are where they left off.
you can also run virtual software at the central end on a single computer to allow more than one host per machine. this actually is a very cost efficient way to go, specially where getting up and going on some other internet platform would take a lot of time to program and setup.
there are other remote programs out there too.

maybe somebody will describe terminal services, the version used today with the new microsoft windows servers.

[MARCO RUIZ VENTURA]

Thank you for the answers.

A detailed explanation of my software .

My software of billing and the database are located in the server, in the client's machine I have created a

shortcut to the application to run the software of billing. The database is developed in SQLServer. The programs

use SQLTools to communicate with the database and use Transactions to upgrade the database.
The programs are a single unit, I don't have module client neither module server.


Spanish:

Gracias por las respuestas.

Una explicación más detallada de mi software

Mi software de facturación y la base de datos están localizados en el servidor, en cada máquina del cliente yo he

creado un acceso directo a la aplicación para correr el software de facturación. La base de datos está

desarrollada en SQLServer. Los programas usan SQLTools para comunicarse con la base de datos y uso Transacciones

para actualizar la base de datos. Los programas son una sola unidad, no tengo módulo cliente ni módulo servidor.

[Shawn Anderson]

I'd say your most viable options are to:

1) completely re-write you program as a web based application
2) use a product like Microsoft Terminal Service or Citrix

#2 would certainly be the fastest to implement but would require a static, public IP address on your Terminal Server. The network overhead is surprisingly small for Terminal Server. It really just comes down to how simultaneous many users you have.

[Mark Strickland]

Terminal Services

Terminal services can solve your problem with ZERO programming. If you don't have a Windows server or don't want to pay for the access licenses look at this:

http://www.thinsoftinc.com/

This will add terminal services to XP. If you read the MS license agreement it is legal.

I have tried it and it works great. The Remote Desktop Protocol (RDP) is encrypted so that part is pretty safe.

My guess is any connection over the Internet (VPN or not) will be much slower than you might think if you are trying to do things like map a remote drive.

The ThinSoft install is very simple and takes only a minute or two. After than all you need to do is make a minor mod to your firewall to allow the inbound connection to be "port forwarded". I am not easily impressed but this is a nice piece of software.