Slashdot Mirror
How Not To Sell Linux Products
An anonymous reader writes "Roblimo looks at why so many Linux products fail in the marketplace, and decides it's not because Linux users want everything free, but because most products they're asked to buy are either poorly marketed or don't work well. He has some good advice for anyone trying to sell stuff to Linux users, except it really applies to *all* computer products, not just Linux." (NewsForge and Slashdot are both part of OSDN.)
Well it's the obligatory SCO Linux license of course!
It's so true! All the linux products I know of (and I don't know of many.. hence the marketing problems) are all targeted at the geek community.
This is not a very large market, and we're the pickiest of users, mostly because each of us thinks we can do it better.
Online Starcraft RPG? At
Dietary fiber is like asynchronous IO-- Non-blocking!
Just today I ordered some herbal viagra and a salve that will increase my memory, products I would have never known about were it not for helpful emails sent by well-meaning strangers. Perhaps people like me could be told about open source operating systems by similar methods. Perhaps the .iso installation files could be sent as attachments.
The flag just makes more sense than the constitution. - Judas Gutenberg
If everyone stopped selling Windows products, and sold Linux products instead, Linux Product Sales would increase.
"because most products they're asked to buy are either poorly marketed or don't work well"
Christ, that's usually why ANY product fails.
Is a lame product like Walmart's Linux PC it?
Or is TiVo it?
Linux products are all over the place, usually concealing the fact that they are based on Linux. Just because Linux and its standard UI are not popular in consumer devices doesn't mean that Linux itself is not used and the products based on it aren't successful.
I have been pwned because my
....his experiences are common to *most* products sold, regardless of underlying OS. The thing that is specific to Linux/geek is that we see no docs and poor installation setups as a fun challenge and brag about it when we conquer it.
....because no apps have ever sold well for linux, as it holds a tiny share of the desktop, there is no incentive to make the apps of the highest quality.
One thing that may change this...I wonder how different OS-X applications are from gnome/KDE apps. Certainly if the vendor uses Qt there is not much difference, but what about using the native toolkits?
The reason I ask is...so many multimedia apps are being ported to OS-X, and Mac users (especially multimedia types) demand stability and dependability.
As more windows apps are ported to OS-X, many by vendors who swore they would never port to unix or linux, is there any chance of these high end apps migrating the extra step to Linux?
Yeah, I'm definitely with the author in saying that companies think they are doing you a favor by porting their software to Linux. I think that people could easily fall into this in the past, but not so today. I guess the free programs are just too good. The payware programs have to not only meet the challenge, but will have to receive rave reviews (like here on slashdot) before people will buy it. I guess that's how I am. Maybe that's why Codeweavers is successful.
In "Open Sources," Bob the Red Hat guy responds to the question "how do you make money with free/open source software" with "that makes the assumption that it is easy (or easier) to make money selling proprietary software." (not an exact quote, but it's close enough). I suspect that most software products actually fail in the market place, or atleast will fall into a small niche market. Linux itself is a niche market, and targeting to niche users in an already niche base futher decreases the amount of potential customers. Even if someone has 100% of the Linux market, that's only like, 10% of the total market, with a liberal (not something i am known for) estimation. So i would say taht the real problem with selling to linux users is the selling to linux users. A company is not going to stay afloat vending end-user software only to Linux customers. Even the most sucessful Mac software producers find it necessary to port their stuff to Windows. StarOffice has the ease of instillation and support on Windows, Linux, and Solaris. Adding Mac to taht is no issue. I've made it work on FreeBSD without too much hassle either.
A successful linux ISV is going to have to have Windows and/or Mac versions of their product to keep the company with enough revenue in order to offer the product to Linux users because the base simply is not there to keep the company opperational otherwise unless the product is truely groundbreaking, breathtaking, or has 0 competition and no free alternatives (not bloody likely). Just my $0.02. Take it for what it's worth.
One thing I learnt from working (as opposed to freelancing) is that you need to take into account business value of a product, otherwise it is next to useless.
Most self-inspired products are too heavily biased towards technology, but not enough in the business sense.
I'm curently researching this spam filter, it may sound like a good idea, maybe it even works, but I have yet to see a business sense in it, i.e. how to market it, brand it and add value to the users. Please note that by business sense, it doesn't necessarily mean profit, but a sense for users to actually use it.
I guess what I am trying to say is, most geek-based products are developed based on the developers' vision of the world, but they hardly have a chance to meet up with potential project sponsors, and consumers (focus groups) who are really the persons to tell what should be developed.
Rock that crushes, Paper & Scissors that don't matter.
Those are just the few I've interacted with recently. IBM, Sun, JBoss, and Novell are doing a very good job of supporting, marketing, and selling their Linux-based server products. So there are more and more success stories out there.
But, like the article communicates, we need a lot more to get the momentum going on Linux for the masses. Hopefully, large organizations will follow IBM's lead, and small, open-source based project will look to CodeWeavers as excellent examples. We need more of those guys!
"Adventure? Excitement? A Jedi craves not these things."
One Linux product that I hope is successful is CodeWeavers CrossOver Office.
It's a non-free product that I bought for my debian system, and I've never looked back.
I may get slammed for this, but I really like Microsoft Word 2000 and Excel 2000 (the later products seemed over-featured-- all i need is well made products: like a good grammar checker to correct inevitable typoes)*. Crossover Office allows me to use them seamlessly on my Linux box. I appreciate that quite a lot.
What's more, their version of Wine works really well for a LOT of "unsupported" software-- from character generators for RPGs to "Teach Yourself Chinese" programs.
Getting their product was a snap- paid online, instant download link to the source and to binaries for a variety of distributions.
Good stuff, and, IMHO, a good example of a quality Linux product that I paid for.
*I'm trying to ween myself off Excel to a more robust alternative, but I find the grammar checker of Word very useful for catching critical, but easily overlooked, typoes in technical writing-- I'd miss it a lot. Is there an OSS grammar checker I am un-aware of?
This statement is solely an opinion. Kindly take it as such in all cases.
I remember being really gung ho about Linux and Open Source after trying my first distro: Mandrake 8.0. At that time, Win98SE and WinME were the dominant flavors (WinXP was just starting to come out), and I found that the Mandrake install did a better job of detecting most of my hardware than the MS install.
Eager to support the cause, I plucked down $100 to preorder the Pro version of the upcoming Mandrake distro. "Cool, I'm supporting open source. I'm doing my part," I thought, and I'd even get some of the CD's early for my pre-order. So I ordered, my credit card was charged, and day after day, week after week, no product arrived. And day after day, my emails to the company weren't answered. There were no real announcements anywhere to be seen about what was causing the delay. Finally, after a bit more than a month of this, I finally called the company at my own expense and had my order cancelled. (And even that required quite a run around, as the number listed on Mandrake's site didn't seem to be a direct number, so I had to call a few times to connect with anybody.)
And this is how they treated an eager customer. Hardly the way to treat a paying customer! I sure wouldn't want to run my business this way.
Granted, things are better now, but when your business isn't run like a business, don't expect customers to stick around. -- Paul
OpenSource.MathCancer.org: open source comp bio
At work I'm doing an embedded linux system. I've got a nice little board which has an XScale processor and some other goodies. It's my job to program it to do various things. The embedded linux system that came pre-installed is taking up too much room. I need a cross-compiler so I can build a new kernel and new system for it.
So I take the dev kit cd that came with it and try to install it by following the directions included. No go. It wants an LSB distro and I use gentoo. I hack the perl install scripts, still no dice. Apparently the install disc has rpms debs and tars. And it installs by converting them all to one of the 3. So if you use debian the install scripts converts the tgz and the rpm to deb then isntalls them. Too bad the conversion program doesn't work. Their tech support didn't help much either.
What did I do? I went online and searched for arm linux. Got arm-linux-gcc from an ftp and patched up the 2.4.25 kernel. When its easier for me to do things myself for free than to use your product what am I paying for besides the hardware? Technically I paid for that support and that software and I got jack. Just too many free and non-free linux things do not work. Sure, there is plenty of commercial software that doesn't work. The odd game here and there and such. But if the company behind it isn't fake you can bet that it is going to work sooner or later. With linux stuff sometimes you just don't know.
This is the opinion of a gentoo user, so I'm not bashing linux as a whole. I'm just saying that if you're going to make something, make it work. Just because a geek is going to use it doesn't mean they want to have to go through more effort to make it go. They just want to go through a little effort to make it go better.
The GeekNights podcast is going strong. Listen!
I don't know. After all these years there are too many distributions of linux wearing the OS too thin.
The penguin has been a good symbol, but how the hell are you supposed to use one symbol to market for 40 distributions? You can't. The only thing that came close IMHO was the redhat symbol. Suse and debian are great distros, but it's marketed like a good singer than a superstar. Christ, SCO has done more marketing for linux in general than any distro.
I am a relatively new user. I bought it because I was tired of Win98 crashing and couldn't justify XP on my 1.1 MHz Celeron. The thing that had me was that Linux just runs.
OO is great. Mozilla is great. KDE is great. Gnome is OK, but KDE is better (IMHO).
What really stimies me is the difficulty in getting USB devices to work (uncommonly used things like Palm Pilots...) and the general difficulty in either updating or adding new programs to the system once installed.
Want to make Linux sell better? Stop developing the latest/greatest KDE, and start working on fixing these areas. Once fixed (and idiot proofed), you will have a distro that costs $50 instead of $39, but the added cost will be worth it. Market the bullet proof operations, and the fact that linux will run on anything this side of a PC-AT, and probably could run on an AT if you wanted it bad enough. In other words, market it to the soccer moms and busy single parents who can't afford to not have a computer for their kids and yet can't afford to pay $1000 for the P4 and $200 for MS Win XP, and the $450 for the Office suite. (I can see the ad now, two harried moms with computers. One has a Tux sitting next to it and one has a blue screen on it. And the caption is "And I could have spent HOW MUCH less?")
t's hard to say: Are the free programmers gifted coders, or are the commercial coders just REALLY bad ..?
First, lets ignore the fact that there is not a clear disctinction between free and commercial programmers. Much of the free work is done as charity by commercial developers donating their time or organizations donating their commercial developers.
Free programmers gifted? No more, no less than any others. What they are is free. Not "free" as in speech, not "free" as in beer, but "free" as in not answerable to anyone else. No boss screaming about deadlines, no my company is screwed if we don't ship by this date, etc. They are free to take whatever time they need, their customers have no financial control over them. Ironically this control often leads to rushed jobs and lower quality.
I'll emphasize
A N D
the dopey stuff of Basics:
Installation Testing
Feature Testing
Usage Testing
If you can't install it striaght off, and start working (either straight away or doing the tutorials... and YOU DO HAVE TUTORIALS... DON'T YOU?) right then and there, you've just blown thousands of man hours as thousands of users bblow their time trying to puzzle out your spaghetti code - and it doesn't matter if it's running in Linux, Windoze, or OSx or whatever. Either it works straight up or it doesn't.
The problem is, WAY too many shops see QA as a an after thought if it is thought of at all, and given the geek-centered history of Linux, it is (sadly) far too common in Linus ware.
One of the main differences between really stunning software and crapware is that the stunning software has a crack QA team running a tight shop with the engineers, and the engineers accept and respect the opinions and findings of QA, just as QA knows the exigencies and limitations of the coders. The crapware has zero QA or the QA consists of the programmers doing basic unit testing, which is too often close too useless due to external dependencies and doesn't address anytihng about UI design...
I did blackbox QA for a very long time, (and still occassionaly do with a good offer) and I have Zero Patience for software that isn't properly tested. Unfortunately, it seems that blackbox is striaght up ignored or sent to India for "chimp testing" (blackbox done to testcases and matrices only) or automated versions thereof, is never brought into the specification process, and in the meantime, it's all gone to whitebox or greybox - which rarely addresses more obvious and critical issues that question basic assumptions in a program, as the lead programmers are too often thin skinned, under served in the social skills dept, and overly identified with the project.
And it has nothing to do with Linux: but the workers in Linux too often have a variable sense of what is an appropriate amount of effort a user should put forth in using a given application or system.
RS
Shoes for Industry. Shoes for the Dead.
The retail marketing of Linux and Linux applications sucks. Of course, all the naysayers are going to declare the retail marketplace dead, but for the general public it's still an important venue.
Walk into any store that carries Linux products. You see some out of date distros. Then you see some new RedDrakE boxes. But what's the difference between the purple Enterprise, magenta Professional and red Desktop editions? There's also FreeOffice in two different packagings, one seemingly generic for a variety of operating systems, and one specifically for RedDrakE which is more expensive. Then you see a copy of FubarOffice 2004, packaged in a tiny DVD box. What the fsck is that? Obviously it's not big enough to have included a manual. Along side it you see FubarPaint and FubarPro. To add to the confusion, there will be the obligatory "UltraLinux Toolkit" containing nine CDs of nine obsolete distros.
And not to pick on Linux, but if you look closely enough, there will be a FreeBSD and NetBSD, each with two different packagings from two different distributors, but containing the same software version.
Don't blame me, I didn't vote for either of them!
Forget marketing, it is the install/config process that makes Linux products hard to sell. Let's face it, Linux assumes that a reasonably skilled sysadmin is sitting at the controls. Windows assumes that the sysadmin can be relied upon to click "Yes", "Next", "OK", and "Finish".
Package installers go only so far. Progress has been made regarding dependencies and cascading installations, but I see room for improvement. I find many products still require the "./configure ; make ; make install" method.
The people who write the code are hard-pressed to consider every possible Linux distro or hardware/software environment. Poor documentation doesn't help. We get away with it on the server side, but this will not work with embedded systems or desktops, where you don't have a sysadmin ready to hack the install. If I am buying a product, I expect the install to be smooth and trouble-free. If I have to sit and hack, I might as well stick with free stuff or write it myself.
Don't get me wrong, Linux products are great, once they are installed. Proprietary products are easy to install, but it's all downhill from there.
I'm a sys admin for about 20 Gnu/Linux servers. The reason I don't use proprietory software is, not because the company can't afford it(we certainly don't mind paying programmers to write for us). The the open stuff is so much better because.
./configure
There's nothing like
make
make install
It's so much easier to troubleshoot a missing library or edit some code to fix a problem.
The documentation that comes with proprietary software is usually lacking. But then the most important documentation, the source, is often never available at all.
I'm sorry this guy had such a hard time. But I'd stay away from those all-in-one commercial products. There's a reason why sendmail, samba, apache, etc. have been around so long. They may be diffuclt to install and configure but have infinite flexibility
"LIVE GNU OR DIE."
Ok, die then. You obviously have no intention of living GNU. GNU is about free like speech should be, not free like people wish beer was. Yes, there are free beer versions of most open source software. However, these mostly come from pay software: Linux (Red Hat, Mandrake, Suse, et. al. are all pay software primarily); ReiserFS (Hans collects enough from support and alternate licensing to live reasonably); MySQL; etc.
Promoting a refusal to pay for software has nothing to do with GNU or open source. Read http://www.tlug.jp/docs/rms.html to see Stallman take to task someone talking like you. The issue is not to get software without paying for it. The issue is to be free to modify the software afterwards. You can get binaries without paying for them, both legally (freeware) and illegally (warez). However, they will never be free software in the GNU sense of the word, as they are unmodifiable.
Living GNU means refusing to use software that you can't modify and redistribute. Refusing to pay for it? You are just leeching off the system. Money is the contribution of those who can't code. Money is also the way that those who can't code can influence development. Thus, paying for software makes it more inclusive by including non-coders in the process.
Would it? One of the points of newspeak is to eliminate redundancy-if you have plus, it can be applied to accentuate the negative or positive following word-minus is redundant
Perhaps there are many outfits out there run by geeks with big egos? They believe that they are center of the world and think that anyone who can't install/use their functionally perfect software don't deserve to use computers in the first place.
The cold hard fact is that many average comsumers have problems installing even simple Windows programs and they are the majority, not us geeks.
1f u c4n r34d th1s u r34lly n33d t0 g37 l41d
More like TiVo. Appliances are where (embedded) linux really shines. Look at the Linksys WPC11/WAP11, CyberGuard, and some appliances that do things on a scale not even attempted in the Windoze world, such as InterIM from Deviant Technologies, and you'll see prime examples of why Linux and other open source technologies are kicking the shiny metal ass of proprietary products.
;-) but there are so many variables involved in making a PC a Linux "product" (OS, office productivity suite, printing, sound, network browsing, etc) that it's probably the worst test imaginable.
Walmart can sell Lindows PC's, and sure, they're interesting, but let's hope that's not what people think of then they think "Linux products". The thing is, despite the candy interface, when you do run into a problem the learning curve is too long.
I've used a Linux desktop exclusively for over a year now, and I'm happy with it, but when I tried to get my wife (a former IT guru) to adopt it it was a total flop. Admittedly, Debian is not your best intro to desktop Linux
Appliances, competing in well-defined niches, are a natural for Linux and they tend to beat their closed-source competitors. THAT'S what I call a "Linux product".
-hp3
So, on the one hand, you have OSS stuff that works *great* but takes *forever* to make and isn't exactly what people (think they) want at any given moment.
On the other hand, you have salespeople feeding people crapware produced over the course of a few months to satisfy the latest buzzword-driven market-craze.
And, suprise suprise, the buzzword-laden focus-group-created crapware wins in the marketplace? This doesn't sound like anything new to capitalism, or anything unique to the computer industry for that matter. It sounds like a *much* larger problem.
"I assumed blithely that there were no elves out there in the darkness"
I often look for software solutions in the open source community, or "semi" open source, like stuff buiilt on PHP/MySQL. Most of the time, I'm lucky if I can even figure out what the damn product does, much less match it to my needs. Basic marketing is so simple I can't understand why bright people can't seem to get the hang of it.
I've got news for all you anti-suit types: Marketing isn't trying to BS someone; it's explaining what your product does, who you've designed it for, and what unique qualities make it better than other choices. GQ's and OB's: Good qualities and owner benefits. If you develop programs and can't do that, you should get a job parking cars or something.
This is not rocket science and it's not hype. It's educating your customer, which is good for the customer and good for you.
--Hi. I'm in Portland and it's raining. This appears to be a permanent condition.
The first product I ever bought for Linux was VMWare, so I could test my Java applications on a Windows platform with relative ease. I think a lot of "cross-platform" type products for Linux would sell quite well. Like a emulator specificly for games, with a guarentee it'll play 99.9% of the Window games on the market. That would be a huge sell.
:) He's very happy to have his KDE interface and Gnome Stones back.
And today I installed SuSE on my machine I'm building for my four year old. I bought the professional version of it for $80 at Best Buy, and was blown away. It was the easiet install of any OS period.
The two manuals are beautiful. It comes with six cd's and a DVD with everything the six dics have. Talk about going out of your way for the customer.
Why Linux for my son? I first had Gentoo Linux on my machine, but had to go back to XP for work related reasons. He hated Windows.
Josh
"Meanwhile, we've found a free software package that is supposed to do the same thing as this unit -- plus act as a print server -- and requires only a minimal computer and a wireless card. We're going to try this method of achieving the same results. It will be scary if free software on a sub-$300 PC is easier to set up than the $1,500 box, won't it?"
Just kinda' curious as something like this may solve a lot of my issues with my small business.
"The strong will do what they want, the weak will do what they must."
-Thucydides
Maybe I'm not a real linux geek, but...
I *hate* when something is difficult for no good reason. A 'monumental effort to get where others have gone before, and anyone with enough time can get' is a monumental waste of time.
RANTThis isn't like the NYT crossword puzzle. The point of the install is not to just have done it. What's so fun and challenging about wasting orders of magnitude more geekhours installing than the documentation/packaging would have taken?
It often seems that developers are masochists. It is not reasonable making people play Where's Waldo with the source just install stuff.
While it's not cool to complain that your free widget wasn't a good enough free widget[ unless you're gunna do something about it]- That only works for those already invested in it.
On the other hand:
1. There are negative consequences to crap doc/packaging. Make it clear that it's less probable to get this thing running than your '64 Fiat "thats been sitting a while." 2. If nothing exists that installs reasonably easily, on most new base-distro installs, then 'Linux' doesn't really "have it" yet. If most people that want "those features" can't get them, you're gunna turn off more potential friends that simply didn't have the time 'you' wasted, than fans who will pick up the reigns. 3. Developers should put a little more effort into doc, maybe cut that 2 day install down to 4hrs. It's far more likely that fans would contribute to building the doc that brings it down to 45 minutes./RANT
Finally, a heartfelt THANK YOU for all the great and not so great FREE software. And, oh yeah, I still brag about it...
No, it's because so many developers (and this applies to small/amateur developers on the whole) focus on skins, supporting skins, creating ghastly skins, skin ranking systems, user-submitted skins (often even more ghastly), and anything related to skins, all of which are entirely irrelevant for almost all software.
Case in point: SpyBot -- brilliant piece of software that I downloaded recently. However, why should there even be "new cool Skins" for a little application that removes spyware from your computer?
'Thats they exact same thing a banana wrench monkey.'
When all you have is a hammer, everything looks like a skull.
Does anyone remember the fiasco with Corel's Wordperfect. I started using wordperfect before MS word was around, in the good old DOS days (sarcasm). I liked Wordperfect a lot, and stayed with it as much as I could, even through all the changes in ownership and the destructive "enhancements".
I was very excited to hear that Corel would port it to Linux. I was a little weary of the Wine hooks they said, but I would give it a try. I paid over $80 for it and what a piece of crap that was. It would constantly crash and I would always be losing data. It would sometimes crash when I tried to save, and the save would lose data or just corrupt the entire file. I finally gave up with it and bought Star Office.
Then, later when Corel gave up on Linux, I read that Corel is an example that you can't make money porting to Linux. I was so angry at reading that, since the real answer was that you can't make money porting shit to Linux. I think Corel expected the "we are doing you a favor" reaction and everyone would buy it. It actually worked with me since I did go ahead and buy it, but I wouldn't buy something else after that unless I knew it worked. I've seen Star Office previously in action, and that was why I later bought it.
Steven Rostedt
-- Nevermind
When you look at commercial software that is made for Windows, for example, most of it is packed with a large number of features that are invariably never used by most users - MS Office, Norton Utilities & Paintshop Pro, for example, are all feature-rich applications but I guarantee that probably only around 5% to 10% of the user base of each one uses the majority of the features that are provided in the software.
The mentality of many UNIX & Linux users is to streamline & optimise their systems as much as possible - therefore, there is perhaps a tendency to veer towards shell-scripting to combine simple tools into powerful programs, rather than using complex packages with features that will never be used.
Add to this that many of us in the UNIX & Linux community (myself included) get very "anal" about optimised code compilation and don't like installing tools that don't give us the source code to play around with.
In summary, it all boils down to the "chicken and egg" situation. Until you get to a stage where you have a large Linux userbase that is reliant on (invariably) GUI-driven commercial applications, no software company will port those applications across - likewise, why port applications to Linux if there is no great demand?
Gentoo Linux - another day, another USE flag.
The main point of the article was not that Linux sucks, but that many companies "do not get it". I.e.
- they take any old trash,
- port it to Linux,
- hope that it sells just by virtue of it running on Linux,
- and if it doesn't they go back and wine in their corner about the cheapness of the Linux users (rather than looking at their own mistakes).
- ===> it's almost as if they wanted to have their Linux product fail, so that they can go back to Windows, and tell management "see, Linux is not yet ready for prime-time"
Yes, the article stated several times that often free Linux products are better than some of these commercial "Linux" "products". This is hardly a "Linux suxors" message, on the contrary.