Graabein asks:
"I'm part of an effort to startup a VoIP provider. We've decided to use Open Source Software wherever possible. Production is not a problem, we can handle the VoIP network itself, POTS termination, web sites, email systems, all the usual stuff. The business side of things is another matter entirely. We need to be able to handle Customer Relationship data, manage subscriptions, handle invoicing and accounts, have a web shop of sorts, online billing, credit card transactions, and more. Whatever system we use has to be able to handle national standards for accounting, or at least be possible to modify to do so. We've looked at Compiere, but our business types are not impressed. Neither am I, for that matter. Requiring an Oracle license is one thing (database independence is 'in development', but it has been for a long time, with no discernable progress), not working properly with Mozilla is another (you need IE to use it fully in HTML mode). What other options are there?"
"Our business types are full of suggestions for supposedly excellent and well suited systems, however they all have in common that they require Windows on the client. If we choose one of those systems our OSS policy is pretty much moot and OSS has been relegated to (some) servers in the computer room and that's about it. I don't mind running these business functions on a Windows server if that is the best system for the job, but having to run Windows on every client in order to access the data is simply not acceptable.
We want Linux and OpenOffice on every desktop. We want to be able to access customer data from a variety of clients, even including Windows. The same goes for Accounting data, HR data, QA data, you name it. Do we have to write our own system from scratch? I'm not sure that is very realistic."
Slashdot Mirror
Phprojekt is another good one to check out too. I've used it for a few small time projects and most people have found it relatively easy to understand. The thing I like most about Phprojekt over OpenGroupware is the install docs are much easier to understand, and for the most part it's worked straight from the tarball.
---- Fight to protect your right to keep and arm bears! ummmm... ya I think that's right....
Your internal IT should never ever never be a gating item for letting your business department do what it needs to do. If the chairman of the board likes MS Word and just doesn't "get" Open Office, then the amount of his and your time that you burn trying to show him the light will forever outweigh the cost you would have paid to get him a Crossover license and a copy of Word and keep him happy and concentrating on what he is supposed to be doing.
www.HearMySoulSpeak.com
True, very true. However if you have let's say an IT budget of only $5000 and you have to get enough machines for 4 people, what then? True you can "suck it up" and purchase a machine with windows installed, but if you choose to use linux as we all know that saves you a few bucks now. Writeoffs only happen at tax time, not at startup.
---- Fight to protect your right to keep and arm bears! ummmm... ya I think that's right....
VMWare $280
WindowsXP License $180
Quickbooks Pro $300
Not risking a business to save $760, priceless.
In the end you'll get more out of doing it right the first time than you will by screwing up your accounting/etc and hiring someone (or wasting someone's time) to fix it.
You're not really asking for a flame, but I think you're disinsightful.
Open Source developers do business as well. Many developers run their business of OSS, or create OSS outside of their work time. Of course, when someone CREATE a business package, they'll have to know what to do with it. But I would agree that in some cases, the searching user would need some insight in what a product needs to offer.
What would customers know? Do you advertise on your enterprise site that you use this-and-that Inc. Accounting Software? Besides, OSS isn't insecure by default, by all means. And, in many countries, like Norway (mine), you own your own information. If a business f**ks up handling your information, they're up in their knees in lawsuits in no time, if users want that.
Bottom line, Open Source is Open Development, not Open Access.
Simple answer: what you are looking for does not (yet) exist.
There are a lot of fancy applications on the net, none of them any usefull for your purposes (and please prove me wrong, I'd though I had been pretty thorough)
Having looked at the same problem for my own small business I'd say that if your business is essential to you, you either start asking for quotations for companies that can deliver a solution to fits your purposes or find a stock application that does most of what you need. (and does it in a way that most members of staff understand it)
Look at the price, and see if its matches your needs and budget.
As you are setting up your own business, you should NOT be fooling around trying to recreate the wheel; you will need al your energy to focus on your business and hope that it doesn't go belly up.
One sure way of doing that is having a dozen incompatible systems hide all your major business information from you, your customers, and your staff.
Don't count on a "web interface" equalling "will run on any platform". I've lost count of the number of "web enabled" applications that only work in I.E. Some vendors seem to think the purpose of a web interface is so you don't have to install new software on your Windows PC (giving them the benefit of the doubt vs just plain laziness/poor qa), whereas it _should_ mean it's platform and browser independent.
Personally, if it will only run on I.E. in Windows (wine/xover office notwithstanding), what's the point - may as well run a Windows app.
If a company goes to the trouble of making a web interface, it ought to be done "right", so any web browser that follows http/html standards can run it. It's not _that_ hard to do.
You will likely need windows for some things, unfortunatly. Fortunatly Wine works very well for a lot of window programs, and since you are looking for which one you use, you can demand Wine compatability from the start.
Don't be a jerk instisting on all open source, you have a buisness to run, and that means spending money once in a while. Don't waste your money (except by sending it to me....), but don't be too frugal either. If you can only get what you need from a pay software, buy it and get on with your buieness.
P.S. buy Crossover as your wine implimentation, those guys put a lot of support into wine and should be helped. (Or alternativly you can get WineX, but they focus on games so I doupt you care about their advantages)
If you are a startup looking to get into the VoIP market, chances are almost all of your customers are going to be running some kind of Windows based computer.
While I applaud open source and use it myself wherever I can personally, and in our offices, we still all have Windows machines on our desks.
If 95% of your customer base (and honestly the number is probably higher) is using Windows to either use your product, learn about your product, or do things like manage their accounts it is foolhearted to not have that technology available yourself.
Our servers and backend systems all run Linux, and yes it does save us money, but don't handicap your business' already statistically slim chances for success by not using a platform most of your customers will!
Your accountants and your tech staff are totally different people right? If the accountants want to use Windows, and your techs want to use Linux, why not?
I can totally understand the desire to be in total control of the software on your mission critical VoIP system, and Open Source makes a lot of sense. But forcing accountants who know zip about it to use Linux is foolhardy; the time wasted fumbling with an unfamiliar system will dwarf any savings (financial and spiritual) gained by using some open source thing.
...we get it all the time, customer wanting xyz because it's the latest buzzword, or their friend uses it or they've been seduced by an evil marketing-droid.
If it fits in with what they need to do and will give them more bang for their buck then go for it. However, sometimes they don't realise that solution xyz has problems efg and that actually solution hij would not only alleviate those problems but cost them less to have supported.
Hopefully they will eventually come around to the fact that they didn't know what they heck they were doing when they specced xyz, that you are indeed the expert and ask for your assistance, net result? Everything will be right as rain.
Now if they won't budge on wanting xyz, and it will be a PITA to support, you have to ask yourself:-
how much will it cost me to offer that support?
and: how much business (on top of the current project) will I see as a direct result of taking them on?
If it will cost you more than it will bring in, it's time to either outsource it or let the customer know you can't do it for the price they want. They'll probably thank you for your honesty and come back to you when everyone else says the same.
I am NaN
So, I'm going to start a business with 5 employees, including myself. I have a $5000 IT budget. I am the only one in the office with any computer experience, which is as follows:
- About 15 years of 'practical' computer experience.
- About 5 years professional experience as a desktop jockey and, later, as a Windows sysadmin.
- I've installed a half-dozen UNIX (mostly BSD) servers for very, very small web sites, but never as a file/application server.
- I've tried Linux on the desktop a few times, but gave up after a week of fighting with any number of typical desktop Linux problems (hardware support, package management, etc).
So, how am I going to spend that $5000? Should I become embroiled in trying in vain to set up FREE Linux/BSD desktop and server machines without any real experience? Network shares, proxies, client applications, network printing -- that would take weeks. Should I hire someone to do it for me at a ridiculous cost, not to mention the support costs for when something goes wrong in my 5 station LAN? How many operations that size have you seen that can afford IT support costs @ even a measly $70 an hour? Or, should I just eat the software costs of Windows, have the entire small LAN running in a weekend by myself, and essentially forget it all exists? True, buying Windows means I can't afford the same hardware horsepower, and I may be stuck using a PC as a server (instead of a 'real' one), but I can fix 95% of it myself.
This happened in countless scenarios that I've personally witnessed (after having been brought in to take over the maintenance once the business gets busy enough). This constant "linux is cheaper" chant is completely, yet unsurprisingly, ignorant of several factors above and beyond the actual purchase price.
-
Inventor of the term 'pardon my French'.
IMO, browser-based interfaces lack in the usability front. They are simply not rich enough.
They work fine on the web because they are a compromise: I give you a site with a rather dull interface , but you get to it without installing custom software AND ALSO you are presented with a familiar and simple user interface (click links, scroll pages, fill up form fields, submit info), so you can catch on quickly with my site, because it work quite similarly to other sites you have visited before (an important aspect that some flash-based and some overworked DHTML-based sites seem to overlook).
Of course, the software used internally in a business has both more demands and less limitations.
It has less limitations because you can install wathever software you want (you have tech support, and don't depend on the end user failing to install the latest plugin). You don't have severe bandwidth limitations. You can standarize on a single platform for your clients, and on a single screen resolution (or, if it is not single, at least can be a known and definite set). You don't need to engineer your application to be run in a restrictive security sandbox, so you can have full control of the devices attached to your computer. For instance, you can make the application print an invoice as part of a transaction, without explicit user intervention (no "print" dialog), automatically selecting certain parameters (paper size, margins, resolution), not allowing the user to mess with them. The application you build can also have a steeper learning curve, because you won't have casual users (potential customers that have to figure out how things work on their own, and that you'll loose if they get annoyed), but permanent users that are your employees and can be trained.
The user interface of business software has higher demands too. If you fill in an online purchase form twice a week, you can put up with a clumsy user interface. But if that is your job, and you process purchase orders from 9 to 5, you'll need something better than the average HTML form. For instance, when a customer tells their name over the phone, you type the first three letters and a list of those customers that meet the citeria is instantly displayed for an easier selection. Also, you might want your text to be spell checked as you type when fou fill a text area. If you have used both SQL-Ledger and GNU cash, or PhpMyAdmin and mysqlcc, or any web-mail and any mail program, or groups.google.com and any newsreader then you should know what I'm talking about: even the best engineered web application falls short to almost all rich GUI applications.
Of course, in the future web interfaces might evolve to become richer (XForms, for intance), but until then, selecting a web-based architecture for internal business use certainly can hurt productivity.
Having said all this, I must also point out that it depends on what you call "a Browser-based application". I have taken for granted that the original poster meant a HTML-based application as opposed to, say, an application consisting in a single page containing a java applet or ActiveX control.
So I suppose you *could* say "It's more important for everything to be OSS, I guess we will just live without a [blank] system", but I'm not sure that's a valid business decision anymore.
You have created a false dilemna. You suggest there are only two options: non-OSS or nothing. That simply isn't true. All your talk about egg timers and burning money is also unwise. It is fairly obvious that the original poster should not just grab QuickBooks while s/he is at Walmart because it is convenient. Nor should a startup jump in with PeopleSoft or SAP or GnuCash, or anybody without a reasonably thorough investigation. Two or three days of waiting for Slashdot suggestions is a minor effort in the larger research.
The original poster is leary of building from scratch. I think that should be re-considered. Every off the shelf solution involves considerable "business process re-engineering." In other words, you have to force your business to work the way the software was built. Universities adopting Peoplesoft have found that they could not pay their faculty "9 over 12" because Peoplesoft wasn't built that way. Professors accustomed to working 9 months, but receiving paycheck all year long suddenly are forced into larger paydays for 9 months, then nothing for 3 months. Don't like it? Pay a lot of money to have the system modified, or do it yourself.
Ask Chevron how much in house programming it took to get SAP to allow vendor payment authorization the way Chevron had always done it before. Ask Fisher Price how many in house programmers it takes to get Keenan/Arbor to handle sales promotions that Keenan/Arbor didn't envision when they built the framework. Ask your local utility how much work it takes to tie your billing and provisioning together. See if they've even achieved it! It's pretty tough when you don't own the source. Oh yeah, and watch what happens when your vendor of choice releases an update to one of the modules. It wipes out all the "fixes" and customizations you have added. Don't cry to them. They can't possibly support every customer's customizations.
Now go ask Verio how five programmers can build a system from scratch that includes ordering, provisioning, and billing with global currency and pricing support in an amazing way that fits the company perfectly, while 200 hired experts struggle on to get a name-brand shrink-wrapped solution out of the planning stages. Granted, five full-time developers isn't cheap. But I can guarantee that no matter what you choose, you will need a team of developers to make it work. And when the solution is not homegrown, that team has to include highly paid outside "consultants."
Build your own system from scratch to fit your company. Don't build/rebuild your company to fit your system.
Oh, and about needing it to work on clients of many platforms... One word: Mozilla. It's a super nice client. From custom XUL apps to n-tier web apps, Mozilla is awesome! And it runs on more platforms than I can name. Of course, there aren't too many proprietary systems out there that have taken advantage of the platform advantages of Mozilla, but did I mention that you can build you own system?
Or... you could just outsource the whole thing to India.
Let's look at the situation.
You're not qualified to administer a Linux environment. You probably don't want to bet a business on it without additional training or help.
But you do have experience with running a Windows environment. So you'll be able to handle that. Your choice will be pretty easy to make.
Of course - plenty of buisnesses consist of people without any IT experience on any platform. These folks will either need training or hire help. And in this day and age, finding help with Linux is not so hard.
The pitfall small businesses run in to is thinking that since they've used Windows at home, they can also manage to run a reliable Windows-based infrastructure at work. And sure - they may get it running at first. But they inevitably run in to a situation where they need to hire help. So much for avoiding the cost of hiring IT experience. This is the scenario that I have personally experienced (and been hired to handle) numerous times.
Whatever you do in the end, don't make the same mistake that tech-oriented people always make. Namely, putting the technology ahead of the business. There is no point of using OSS just because you want to. What comes first is the business. What is best for the business? If it is Windows, that's what you should go with. If OSS works out better, that's what you should use. Also, don't forget that you can have a mix. For example, you can use some Windows software for the business process tasks (say CRM or something) but use linux (openoffice,etc) for basic desktop use. Depending on what you need, you can pull your customer data from the Windows database (say MySQL, or MS SQL Server, or Oracle) into a linux application (this depends on what your final business software allows).
Sivaram Velauthapillai
Sivaram Velauthapillai
Seeking the meaning of life... @slashdot of all places
No - doubles are just as lousy, in the long run.
Use Int64s.
$92,233,720,368,547,758.07 to -$92,233,720,368,547,758.08 should be enough range for most folks. Most governments, too.
Education is the silver bullet.