Slashdot Mirror
Key Advantage of Open Source is Not Cost Savings
cmcsonar writes "Computer Economics recently conducted a survey of visitors to its website regarding the perceived advantages in the use of open source software. Although not a scientific sample, the results are nevertheless startling."
Indeed, Microsoft may be find for small systems and hobby use, but when you scale up proprietary software becomes a worse and worse choice. I had an interesting experience in a startup company when Yahoo was considering to buy our company. They sent a bunch of people over to review our technology; and when we mentioned our databases ran Oracle, one of the guys looked to our CFO and told him "you shouldn't have let them do that". "Why not", our CFO asked - since he was one of the guys orginally insisting that to be taken seriously we'd neet to have top-tier components everywhere. The Yahoo guy's response: "Well, Oracle may scale well technologically, but it doesn't scale financially".
If you could see what my customers have been through with vertically targeted vendors, this would be obvious.
I have one customer who paid $30k for a point of sale system (5 terminals), was paying probably $5000/year in reseller support costs, etc. And it isn't that great of a system! Furthermore the vendor will only support the reseller who originally sold the unit, so he is married to them for support.
Not so with FOSS.
LedgerSMB: Open source Accounting/ERP
(It's not karma whoring when you're mister AC)
As nearly everyone knows, open source software is a low cost alternative to proprietary software. For example, the open source Linux operating system is commonly seen as a low cost alternative to Microsoft's Server 2003 operating system, or Sun's version of Unix. The popularity of open source is seen in the fact that today the largest market share for web servers is held by the open source Apache system.
One might think, therefore, that the key advantage of open source software is its low cost of ownership. But visitors to our website didn't think so.
Open Source Does Have Advantages
Our survey offered respondents a choice of five advantages for open source.
Before we discuss at the topmost advantage of open source, let's look at what respondents are not saying.
Even though advocates of open source products such as Linux tout its security, only 3% of repondants ranked "higher level of security" as the key advantage of open source in general. In addition, although open source software is by definition open to user modification, only 17% of respondents ranked "easier to customize" as the key advantage.
Furthermore, only 14% of respondents thought that open source had no significant advantages over proprietary software.
Free is not free
So, what is the top advantage of open source? The leading vote-getter was "reduced dependence on software vendors" at 44%, followed by "lower total cost of ownership" at 22%. Although these were the top two vote-getters, it is enlightening that respondents valued reduced dependence on software vendors by a two-to-one margin over lower cost.
The second place ranking for "lower cost" indicates that IT decision makers recognize that open source software is not really free. With most types of software, administration and support costs overshadow initial software license cost and annual maintenance feesthe costs that are minimized by open source. Therefore, software buyers do not see the low or zero initial cost of open source as its most important advantage.
Whether open source software is less costly to administer than proprietary software depends largely on a ready pool of resources trained on the system, the availability of administration tools that allow system administrators to manage a greater number of systems, and the number of version upgrades and patches that are issued by the developer. In this regard, open source software may have little if any advantage over proprietary software, although the situation varies from application to application. Therefore, low cost, although important, is not the key advantage of open source.
Valuing independence
The survey indicates that IT decision makers value "reduced dependence on software vendors" as the most important advantage of open source. This indicates that software buyers must feel some level of dependence on proprietary software vendors, from which they desire freedom. Such dependence includes reliance on the vendor for maintenance and support and the necessity for the buyer to accept version upgrades that the buyer may not need or want.
For example, when Microsoft announces a new version of its Windows Server operating system, it invariably phases out support for older versions of the system. Users that are satisfied with older versions of Windows will be eventually forced to upgrade if they want to continue receiving vendor support. In contrast, there is no forced upgrade cycle with open source. Older versions of open source products continue to be supported through the open source community and third party support providers as long as there is demand in the marketplace for such support.
Our survey indicates that vendors of proprietary software are missing the mark when they argue that open source software has a higher total cost of ownership, is less secure, or higher risk in terms of ongoing support. These factors, although important, are not the key concern of software buyer
I've been saying this to everyone for years! The reason I use open source is always because it is BETTER than the other options.
Here's some examples:
CDex > Real, etc.
Firefox > IE, etc.
Gaim > Trillian, aim, msn mess, etc.
thunderbird > outlook, etc.
pj64, 1964 > real n64
shareaza > kazaa, etc.
VLC & MPC > wmp
phpbb > vbulletin
etc. etc.!!
The only commercial products I still use are:
Winamp. It's the best!
Nero. It's the best!
Dreamweaver. (N|vu is getting closer! Fix the table selection code already!)
bsplayer. It's really good for video!
Civ 2. Freeciv is still ugly and clunky!
And that's pretty much the whole list of what I use.
I hold very few opinions. I hold information based on observation and fact. If you wish to disagree, please use facts.
1. Integrates with back office financials, such as Quickbooks?
:-)
Why would it need to? SQL-Ledger is more than capable of doing the accounting. Take a look at it: here
2. Handles integrated credit card processing... through with merchant services?
Forthcoming. Have a customer who just agreed to pay me to develop this feature.
3. Handles all standard POS hardware, such as card swipes, receipt printers, pin pads, cash drawers, pole displays, and touchscreens?
Touchscreen (keyboardless) support forthcoming (assuming my current lead pans out for sponsorship of this feature). Card readers, barcode readers, pole displays, receipt printers, cash drawers, etc. are currently supported.
Honestly, most of this is *really easy.* Barcode readers are currently only supported as decoded signals via keyboard wedge interfaces. Same with mag card readers. We support the Logic Controls PD3000 pole display, but others could be added with minimal work.
Printer redirection is also supported btw. So if you want all your restaurant orders to print to a printer in the kitchen, this is relatively easy to impliment.
For example, one of my customers is using a computer with insufficient RAM as a terminal to the main server. The printer and pole display redirection allows them to redirect it to the terminal even though the application is run via firefox on the main server!
4. Can be easily taught in less than 5 minutes?
Yep
Well, the online cash till countout report might take another 5 minutes... But that is currently being worked on.
We are currently working on resolving the last three critical bugs before release.
Honestly, the currently supported features so far cost approx $2000 in development time. This is nowhere even near the high 5 digits you speak of because SQL-Ledger, while its architecture (CGI) may seem at odds with the environment is really solid, and can be easily extended to provide for all this.
Forthcoming features (ones that people have agreed to pay for at the moment) include:
1) Portable data terminal support for receiving and inventory adjustment
2) Cashier-based roaming till.
3) Integrated credit card processing
4) Movie rental application
5) Coffee card discount app (as in every 10th espresso is free of charge)
6) Credit card processing with a DBI/DBD-like model (allowing for easy expansion for additional gateways).
Likely features within the next year:
1) Keyboardless operation
2) Restaurant module
3) Bookstore module (online ordering, books in print, etc)
4) Online shopping card module.
LedgerSMB: Open source Accounting/ERP
For pointing out the quality of OSS I would have picked something that was actually OSS.
Thats just me though.
I watched a reverse sales-pitch from the CIO of Massachussets (Peter Quinn -- nice guy, sharper than most engineers I know in industry) basically pleading for there to be more OSS available for e-government. Their #1 worry with MS was lock-in, that they'd save formats in .doc in 2005 and then be stuck buying upgrades to Office 2025 twenty years later just to be able to read their own archives (with constant "improvements" on the closed standards necessitating the upgrade chain). The guy was smart as a whip.
You want to sell software to Mr. Quinn, who literally has a couple million for budget in that department? Be best in your class. Now, a little secret about that: after he installs OpenOffice on his boxen, he's every bit as locked into it as he would be if he installed MS Office. Those transition costs are basically capital and when you move to new software that capital depreciates to 0 almost instantaneously. Training, for example. ./ readers might laugh their heads off, but the vast majority of the government workforce does NOT rate an A+ at their wordprocessing abilities -- they know what they use every day, and that means they know Office. And if you have them use OpenOffice for 5 years instead, they'd know how to do common tasks in openOffice. But switch to a different OSS wordprocessor or back to Office or to a third vendor and all of that knowledges vanishes like dew in the morning sun. When you're talking about large organizations, the disruption in operations retraining causes runs into the millions. Its annoying even for smaller shops -- I've in R&D in a mid-sized government office in Japan and when we switched IMEs (input method editor: the thing which lets you type Japanese characters on a Western keyboard, essentially) half of our secretaries suffered major freakout and lost 2/3 of their typing speed despite the fact that the interface is *identical* (and this results in them calling up R&D confused about WTF happened, because we're not a large enough organization to have a dedicated support staff -- which leads to 6 people with PhDs spending about 30 man hours teaching 20/30-somethings to touchtype).
Help poke pirates in the eyepatch, arr.
Browse through the list of proprietary software packages that got canceled - even from big successful companies (PeopleSoft, HPUX, Bluestone and any other software HP touched, Win95/Win98, Dos, Windows3.1). And add that to all the proprietary software that failed, and you'll see that the odds of your contribution having future value in proprietary software is even lower.
> You can't have your cake and eat it, too.
Linux, FreeBSD, and Solaris all do just that, thanks.
> First you complain that Microsoft has an unfair
> advantage with bundling their apps. You want them
> to be forced to unbundle their apps. Then after
> they unbundle their apps you complain that
> Windows doesn't come bundled with any apps.
Who is "you" here? I didn't say that. But now that you bring it up... (see below)
> I think Microsoft did have an unfair advantage in
> bundling their apps, but I'll be damned if my
> government is going to force them to bundle
> competitor's apps.
More stuff I didn't say. But now that you bring it up...
1. No government forces any of the the OSes I use (not even Windows) to bundle certain apps, so this is a straw man.
2. I never asked the government to do so, either. (See #1.)
3. The difference between what you get in a typical Free OS install and Windows is that the Free OS offers you *choices*. Would you prefer OpenOffice.org, KOffice? (AbiWord? Gnumeric?) Kopete or Gaim? Kchat, Konversation, KSirc, or XChat? KDE or Gnome? (or FWM, WindowMaker, CDE, Java Desktop, IceWM, etc.?) Mozilla, Konqueror, or Opera? Mozilla Mail, KMail, Pine, Thunderbird, or Evolution? And so on. Each of these will do the job, each has its own bells & whistles (or lack thereof).
Il n'y a pas de Planet B.
Modern Linux distros (and other similar) operating systems are easier to install, configure and maintain that the MS variants I've observed.
Maybe for you and me, but for an IT department built around Windows with experienced windows administrators that is definitely not the case. I've seen Nix geeks struggle with basic problems in Windows and Windows geeks struggle with basic problems in Nix. Its all relative and your data point is not true for everyone.
"The words of the prophets are written on the Slashdot walls."
Hello?! The QT toolkit issue was resolved a very long time ago. It's available under commercial and GPL license terms. The horse died a long time ago, stop beating it.
"The words of the prophets are written on the Slashdot walls."
A slightly more scientific survey (slightly) run by IDC (as reported in Techworld) also indicated that price was not the main factor driving businesses to open source. This survey focused on Western Europe, and had a few interesting points, such as: only 25% of the companies surveyed used Linux, but 33% use OSS database products.
One thing I found curious: "industries that treated software as a commodity were less likely to have open-source deployments." Again, a bit backwards from what one might expect. There were also, reportedly, a surprising number of respondents who said that the ability to customize the software was important. This may be related.
Why "bar sell it" ? There is nothing in the GPL whatsoever that prevents you from selling a piece of GPL software for whatever price you can get.
Actually, WinXP doesn't detect my gigabit ethernet card, nor does it detect my on-board SATA controller or drives and about a half-dozen other (but minor) devices - this makes setting XP up a huge pain in the arse, requiring a floppy with drivers for my SATA card on installation and a copy of my ethernet drivers on my USB memory stick post-install (I access the internet over my LAN, so I can't just download them from the Gigabyte site). Debian/Sarge otoh automagically detects all my hardware (including an obscure TV card) and sets it all up for me with no fuss, all I have to do post-install is apt-get the nv binary drivers from the non-free repoistory... Post-install of Sarge, only selecting the Desktop environment gives me the choice between Gnome/KDE, the choice between Konqueror, Galeon, Mozilla Suite for my web browser - the choice between Evolution and something else for my Email client - the choice between Kopete or Gaim for my IM client... on top of this I can simply su and apt-get Firefox/Thunderbird/Dillo and so on as I need. Then when patches are released I can apt-get update && apt-get upgrade and have all of these applications updated - i'm sorry but Windows will probably never reach this ease of use.
There's mischief and malarkies but no queers or yids or darkies within this bastard's carnival, this vicious cabaret.
Eclipe, i'd say that Eclipse for developing Java applications is far superior to Visual Studio for developing any of the languages it supports (although admitedly i've only used it with C++).
There's mischief and malarkies but no queers or yids or darkies within this bastard's carnival, this vicious cabaret.
If it takes you an hour to install Office then there is either something wrong with your PC or maybe the drives are just really slow. A full install of 2003 Pro on my machine from the MS DVD takes no more than ten minutes from start to finish.
It does sound easier when you put it like that but yet again, in my own experience, patching my FC3 box and my Wintel boxes show no clear winner in terms of time to patch, especially when the XP Pro install disc has SP2 slipstreamed onto it.
As the machines are both behind a firewall I can be comfortable that they can both download the updates at night. Depending on the environments, both *nix and Wintel boxes could be more at risk without the patches in place. Look at the Linux security newsgroups to see numerous people getting their boxes rooted.
What happens when a Linux distro refuses to install ?
> You must realize that the moment Windows includes every single "extra" you point out, there will be an outcry of "monopoly!" heard round the world. Last time I installed windows, it came with a browser, email, and IM btw...
Microsoft was't accused of being a monopoly because they bundle their own applications.
They were accused of being a monopoly because their contracts do not allow PC manufacturers and retailers to bundle _other people's_ application.
In other words, they're not a monopoly because of what they _include_. They're a monopoly because of what they _exclude_.
And that's not even counting Microsoft's various other illegal acts, including sabotage ("polluting" Java), fraud (FUDing DR-DOS), contract interference (cutting off Netscape's air supply), and so on.
Now the courts may have reacted by telling Microsoft to unbundle some of their own applications, but they did that in order to remedy the damage done by Microsoft's exclusionary contracts and other illegal behavior.
--Depending in hardware, I guess. 1.5 seems a touch long though...
Driver updates and install 1 - 2 hours
--There are programs (I use nLite), that let you add drivers to the windows install CD
Driver updates and install 1 - 2 hours
--Ever heard of Slipstreaming? You can make an install CD with the updates already on it.
Office Install 30 minutes
--I call bollocks. 30 minutes on a slow machine off a CD, maybe. If you're smart though, you'll be running off a network share or a thumbdrive.
Office updates (who does that?) 2 hours
--Ever thought of keeping the office updates on the network somewhere? But then again, who actually does Office updates?
Securing and configuring all that crap 1 hour
--Guess it depends on how much you're doing. Some settings can be made systemwide through gpedit.msc. Some can be stopped by removing services (like Windows Security center).
Anti-virus 20 minutes to 2 hours depending.
--NAV installs in 15-20 minutes, max where I work. And that's on the older hardware.
As far as using images, if you're using relatively homogenous machines, they'll work fine. Otherwise though, I agree a fresh install tends to work better.
nLite
This is the program I mentioned. Allows you to slipstream Service packs and hotfixes, add drivers, remove components and services, and make unattended installs. Its a huge timesaver, if you're willing to make a new CD every month or two.
He does this regularly. His poll question this week is "Is your organization outsourcing any IT functions to offshore providers?" You can answer it here.
This is probably less meaningful than Slashdot polls. No CowboyNeil option, either.