See, the reason why Microsoft is so popular is also the fact that their killer application is not Office tools, it is VISUAL STUDIO! No one seem to recognize that. That is the reason why there are so many productivity applications on microsoft platforms. Also see how MS products are. Have you looked at Access? How many wizards you can create which helps people in their day today lives? Where are they in Linux? If there is an equivalent of Visual Studio bundled with every linux distro out there, Linux will 'get it'.
Anther de facto thing to have is having WINE, nicely integrated so that most of the MS products will work on Linux platform. If that can be integrated well, there will be a time period when both linux and MS applications can work side by side, and slowly phase out, when Linux applications develop. And RH abandoning desktop, I think on a corporate level, SuSE has the best chance. Esp. when Novell is backing it.

build the desktop around html and svg. continuing to hold onto a desktop as somehow distinct from the web is just stoopid(tm). look at the confusion between Windows Explorer(my computer in newer versions of windows), internet explorer and my network places. this is just stupid. just one interface that gets me to stuff on my computer or brings stuff from my lan or stuff from across the web to a single desktop access point seamlessly would kick ass.

svg is good because it gets around some of the sticky "scaling" problems that many desktop elements based on bitmaps have. on my 21" lcd at 2048x1536 the mouse cursor and icons are impossibly small. i realy my resolution is not the norm,but in 2 maybe 3 quarters from now it will be more normal.

Shaitand -- and several others -- are right. A stable alternative to QuickBooks running on Linux would result in an avalance of distro adopters/M$ refugees. I grit my teeth and use W2k in order to run QBPro. As much as I despise Intuit for their introduction of spyware into last year's TurboTax and Proseries products (I abandoned them after a decade of professional use), if the CodeWeavers can get QBPro to Gold level in Crossover (anything more than "can use MOST features") and get one of the major tax programs (I'm now using H&R Block's products) to work seamlessly, they will see a solid migration to their product (and Linux) even without a native Linux alternative to Intuit's really shoddy products. (God forgive me!) I suggest we lobby Intuit (and anyone else who will listen) to port a (proprietary) QBPro and (proprietary) TurboTax AND ProSeries to Linux in order to accomplish the mass migration. The objective would be for the community to rebuild Rome (i.e., create an Open Source alternative to QuickBooks and TurboTax) at its leisure rather than trying to build the PERFECT UserLinux up front. Remember, we need a good plan -- not a perfect one. If you look closely, the major advertisements on these pages are paid for by M$ -- NOT by any application vendor -- does that give you the willies?

Open Office, hmmmmm? Not to start a flamewar, but I still think that thing is wicked slow. Abiword/Gnumeric are blazing fast, though I'm not sure about feature differences. That being, said since I haven't noticed, maybe that falls into the 90% of users use 10% of the features adage....

The "RPM Hell" problem has already been solved. Conectiva ported apt to deal with RPMs already, even the MD5 sum and GPG signing stuff, which could do with being back-ported into the Debian version.

I use Debian myself since RH went non-free, but still have a few unconverted Red Hat boxes about. They all run RPM apt.

Vik :v)

Shaitland has a good point about a Quickbooks type of app. One unrecognized benefit of that suggestion is that (volume wise) there are more customers of that type app than those needing an OS400.

Make a distro that doesn't break everytime you unplug the mouse.

I like the the idea of a "not ready for Linux" list. I wish I new my new Dell Latitude D800 was going to be such a PITA.

Bruce, I posted this in Slashdot as well. Hopefully, it will reach you somehow.

I think the instincts that have driven you to create a new distribution are very much correct. But let me reiterate a few things that I as a user think are important:

A community distribution that serves as an active and clear implementation of an evolving LSB that both software and hardware manufacturers can focus on.

A community distribution that honors the lofty goals that those working on Linux set out to accomplish. This means no-pear seat licensing, in fact, no onerous licensing terms of any sort. Red Hat or SUSE are to expensive for the developing world and even for small non-profits in the US, simply because they added cost of their yearly support agreements is beyond what they can pay. For the record, I am currently using Mandrake as I can freely redistribute it and the keep their security updates on a distributed network of FTP servers, the way that Linux was traditionally distributed. In summary, it is paramount to have a distribution that commits to keeping security updates for at least three years.

Bruce, don't start anew. Linux is all about standing on the shoulders of giants. So if you can adapt Anaconda or Mandrake's installer to your distribution, all the better. These are good and tested tools. The same goes for configuration tools. Borrow as much as you can. Ark Linux also looks very promising and very integrated.

Software installation is not difficult if you have the correct repositories. Preconfigure this for the user and provide a tutorial that shows them how to add new software. Adjust expectations by telling him that all software will be now available just a click away. URPMI and apt-get are great tools. Make them look pretty a la Lindows and the problem is solved.

The desktop is far more complex than it is made out to be. It's not just about email, office software and mp3 playing. It's about accounting and instant-messaging and multimedia. Let's popularize the ogg format a hell of a lot more. Let's include in the distribution's web site a list of radio sites that broadcast in ogg ( i have such a list). Let's work on getting Realplay to really open up its format as they said they would do with their Helix player. All of these things need to happen.

Finally, I think your distribution should link a lot more closely with Linux true power base: the LUGs. Work with them, talk to them, make it easy for them to promote it. Make it easy for them to be involved. A Pan-lug UserLinux forum would be a great thing. I am looking forward to the day when we can differentiate at a higher level of system design. Distribution differences, particularly on the desktop, are getting old. If you are a successful, you may lead other distributions to join forces with you. At least, I would that the smaller ones, ArkLinux, VectorLinux, Yoper and even Mandrake would.

Suerte.

I believe for the desktop and server(possibly more the server), that what's missing from apt is the ability to request a full system check of known files. Such as, with RPM systems, you can issue a rpm -V , or rpm -Va to check a package or the whole system. This is a sorely missed item in deb based distros. This is a great way of checking the state of a package on your system.

All the features are nice, but what really matters is giving the user minimal or zero system administration, and that's not just on install, that's when they want new software and want to upgrade. That's no small task.

Windows needs to improve in this area, but linux has even more improvement required, and must get much better than windows. To make users switch it is not enough to give them what they had before for free. Most users don't even see their dollars going to Microsoft, it just comes with the PC. (They do see it if they buy Office.)

You must give them more than they get from Windows, and the free is a bonus.

More does not mean more features, it means less hassle.

I got my first computer 13yrs ago to make music(with midi)
in that time I have learnd more about computers/OS mostly
Dos\windows but have Suse8.2 makeing music has become big
on PC. Better support ProAudio would nice. since on my
windows box all have to do is buy a Pro soundcard USB\PCI
install the drivers and software. I just want to make music.

Open source Citrix replacement, maybe a derivation of VNC. This will make it easy for small to mid-size businesses to run Windows-only software clients on any OS, including Linux. The business would only need one Windows machine running the "App-here" server (or whatever the open-source Citrix clone will be called).

The rest of the computers can run any OS that the open-Citrix client supports. They would still be able to run the client app that formerly only ran on Windows. Anyone could deploy this solution since it would not cost the large heaps of cash required by a full scale Citrix deployment.

Voile! Client-side OS lock-in is totally gone!

My only complains with all those distribution (or rather question) is that they all want to take on windows but they are lucking few important application and it seems to me nobody is doing much to improve the situation. And those application are:

1.)
Replacement for Winfax. Hylafax is a good server but nobody want to
write good GUI to manage the faxes the way WinFax does.

2.)
Good accounting package for small business, to replace Quickbooks.

3.)
Native Tax Program for Linux, (not copy protected software from Intuit).

Joseph

For Home Users.
======================================
Should be able to play as many media types as possible(avi,mp3,mov,wav,dvd etc.)

Simple development environment. Something like XUL from mozilla, but works with native desktops (KDE, GNOME, even Windows ()), and simple as possible. Languages will be embedded like javascript is in web pages. Interface for multiple languages like php, javascript, java, perl, python. Hopefully will result in lots of medium/small applications that do lots of things users want (family trees, puzzle games, photo management,setup screens, educational, entry forms(Good for business users to)) , but would be very simple to write, and will result in well written apps. (lots of pre-defined objects).

Removable media should be automatically mounted and unmounted(even if files are open but not active). Link should appear on desktop, or in any specified folder. (I think there is an overlay file system that can handle this)

Business Users
===============
Access replacement. A user level database is needed. Mysql/Postgress don't fit the bill, can't be mailed or copied(by users), are too powerfull/complex. The user level database should be able to connect to multiple data sources. Either Access compatibility or a conversion script would be good.

System management. Ability to remotely install and upgrade software on a group of systems. LDAP support for users and groups to manage across the whole network. Network installs, better than kick start, configuration is on the server, instead of a floppy. Third party software is installed during setup.

Lots of predefined install types
Disk server(NFS home directories)
LDAP server(users, groups, etc)
Workstation(NFS or Coda mounts /home)
Laptop(Coda mounts /home)
Terminal server(X application server nfs mount /home)
Skinny client(X client, / on hard disk, no home)
Thin client(X client, network boot / nfs mounted, or ramdisk)
Some logical combination of the above.

With Xandros, Mepis, Knoppix, UL is getting closer. Perhaps less kernel and program tweaking and more intuitive GUI aides for the inexperienced. Delay the bleeding edge in interest of adding users. Also better help without jargon like apt, grep,cron, and other incomprehensible gibberish words/phrases. Linux would have to be de-nerdified. I comment as someone with no computer training whatsoever and able to self install GUI versions of Linux on 3 computers, but frustrated with the extensive fancy in-the-know geke language and no decent resource to educate myself with and look up stuff.

Clean up CD auto mount/unmount. I usually end up disabling it and doing it manually because automount is too unpredictable, and unmount on eject is usually non-existent.

With Xandros (also built on debian) Linux on the desktop is very close. I've used lots of distros.. and Xandros works. The networking in a windows lan is stellar. It looks nice, has the best programs.. very close.

My advice to UserLinux.. make it work like Xandros, but FASTER!! I like that it works.. but very painful compared to an older version of windows. I still use 98 cause 2000 is too slow.. If we can get a linux to run faster.. and as Shaitand said, make accounting software work on it.. we will have a winner and can really begin decreasing the amount of win doesn't boxes ;-)

What is your goal here, making a complex tool look simple or make a simple tool that can do complex things?

If the latter, look at OS X to set your goals.
A pleasing, forgiving environment that does the obvious without a hassle and gives power and complexity where appropriate and requested. It installs with the bare minimum of tools, but the best of breed - which incidentally all adhere to the same design principles you've enforced on your windowing/desktop environment. It does printing, font management, disk burning and other small and 'simple' things without the user having to have a clue. It talks to hardware and for the most part doesn't require driver-hunting - let alone driver-writing...
It invites programmers to adhere to the design principles - maybe the hardest part of it - and updates itself.
It has command line as an option, not a necessity.
And it looks good, breathes calm coolness without trying. It just is.

If the former, just plunk in some buttons and apps and declare it dumbed down.

Two ways the "latter way" can happen...
1) it's enterprise-driven and paid for
2) Linux will come with your personal avatar who'll shimmer and shine and handle all the chaotic complexity that makes linux so sharming but basically useless for the average user...

For Grandma - a file system simply designed for a single user.

you also need SIMPLE media tools: super-simple digital camera connections and a photo-management app, super-simple cdrw support, (i.e. drag, drop, click burn), DVD playback with full 5.1 audio support, DVDRW burning, and EASY network configuration (i know a lot of the newer distributions have some of this, but it's NOT well integrated and the driver support is horrible)

Cool, great ideas. Get to work.

- A good frontend for software installation and update is a must. Synaptic is a step in the right direction. Also, there should be a feature to search for a program to perform a specific task. Often the right program for the job is already installed in your distro and you don't even know it. This could be as simple as a text search of the descriptions in the packages.

- A good integrated help system. Help in Linux is a mish-mash of man pages, info nodes, and online help specific to programs. You really need just one help system, specifically one that is easy and intuitive to use for the novice (unlike man pages or, especially info). Since much of the content exists in man pages, info nodes, or on the web in other forms, it would just be a question of collecting it and converting it to a common format.

- A complete set of working plug-ins for the web browser that are easy to install. In many distros getting all the standard browser plugins (Realplayer, Flash, etc.) working can be a bit tricky. This should be easy to setup, since it's something most people will use. An extention to this would be setting up software for viewing video (.avi files and such) that can be a bit tricky to setup.

- System configuration tools that actually work. Many distributions have system configuration tools that are unstable, non-functional, or not setup properly. Sometimes they have several programs to setup the same aspects of the system (SWAT, Linuxconf, KDE Controls, and others). It's better to pick one and make sure it will work.

- Generally, start with a simple system that will work well. Things should be setup so that the average user will be able to happily use the default setup without having to delve into the innards of the system but the more advanced user still has the ability to alter the system if he wants to.

The list is simple, but long:
1) Easy no-brain installation, all questions up front so you can start the installation then leave it alone. auto hardware detection. Well chosen default applications for newbies. No button that says 'install everything' (every newbie I know clicks that, then complains that there are too many programs in the menus). Call it advanced or something.
2) Very easy GUI config tools that actually work (and easy to find the tool). Try telling your mother from 3,000 miles how to change dial-up settings. I can never remember what the GUI menu entry is called and editing a config file over the phone is impossible.
3) Very easy patch installation process. (one or two clicks with option to automate. Need to have an editable server list to pick up updates from).
4) Only best of breed (for ease of use) software by default. (Open Office, Evolution, Mozilla, Gimp ...) Choices are great, but not by default, it only confuses newbies. Non newbies can click that 'advanced' button to get other options.
5) An install/uninstall that works, and maintains the entries in the 'Start' menu.
6) Multi-media that works and stays working.
7) Games, but not too many by default. Better to include 6 or 7 great games, than tons of games. Again, avoid confusion, keep choices to a minimum for newbies.
8) Easy remote file synch. (Docking laptops, or personal systems to office over the internet)
9) Easy VPN creation.
10) Make it easy to create a boot CD (with /, /etc, /usr) for normal running and leave the hard disk for data (/home, /log ...) This is for those who don't patch.
11) Make it easy to create an ltsp type x-server (and client).

And it all has to look fabulous.

Gord.

Shaitand: RE: Accounting Software

Accpac has a Linux version of their accounting software. They are actually the ONLY accounting program ported from Windows to Linux, seems to be a good move in my eyes.

I'd like the GUI to be more scriptable, which might convince more CLI folks to take notice. Start with the ability to record macros.
It might be nice to be able to run apps in a perhaps slower mode, that nonetheless allows you to screen grab & parse a glade-like XML description of the current window.
Alternately, a tee of streaming ascii data that I can pipe to my Visual Grep or Visual Sed!
Even better, if I can intercept the display.

Hardware detection.
Adding new programs/updates - make it one or two clicks, max.
Enough with the "productivity programs/browsers and such, make what we have BETTER.

Linux user # 304265

The biggest hole I see, at least for the purpose of approaching the small (when I say small, I mean 5-50 employee's not 100!, we are talking the masses here) business desktop is accounting software. I'm not talking about Quicken, you can do your taxes quite easily right through free services on the web.

I'm talking about Quickbooks and Peachtree, a good accounting package with all the features in these two apps as well as the ability to import and export to their data formats could grab a vast market. This market is BIGGER than corporate even today. It has the advange of only needing the right set of applications and being immediately deployable. Their needs don't change rapidly or require going to Walmart and buying software. Their local tech shop is generally all the "service contract" needed by these businesses. If linux takes this sector it will weave threads into the corporate desktop/file&print server because there are a volume of businesses using it, and enough corporate examples already. Yet these businesses are small enough they will achieve the personal reference impact to thread out into home user desktop acceptance as well.

After all, linux is NOT windows, the community does not promote the average user performing actions which have consequences beyond their knowledge and the high ranking employees and owners of these businesses can afford to pay a tech to come in and install linux right. They can afford REAL computers instead of walmart specials running windows that are tuned as windows pc's in this class for performance. Their relatives work with those computers and gain experience and maybe they end up with walmart specials. But regardless of some of their problems, they won't question linux since after all it runs just fine on aunt dorthy's computer.

If a good accounting app like this existed it would be the final piece to the puzzle that would like us migrate the bulk of our small business customers and their home computers to linux. The only ones left behind would be those like yamaha dealerships that gets windows only parts ordering software and such sent from other companies, even if that software runs on wine now there is no guarantee development on the software won't change that next year.

- Also, put in "training center" into Desktop. that can link users to the center. (play in media format).
- May be have custom function. Let user have choice to active the custom function or not. Users can custom their Desktop's outlook as they like.

- Keep the Desktop as simple as it can.
- Have a software bank (build in from the computer/installer disk/internet) that allow users to select the desire application and active it easily (install it easily). The applocation in software bank is shown as catelog. That allow users to shop in there.
- Everything keep it simple, no more
- Build in support news group that allow users easy to click on and find the solution. Keep it easy. May be a funny icon
- Stable
- Such as IanM said, Sysadmin has the control and de-activate the application

The system user has to be able to install software unconsciously and have the machine take care of itself. For example, the software should be installed separately from the system itself in a fashion that other users can use it if required but are not forced. When a user starts the old program they are informed (once) that a newer version exists and they can use it instead 'with a revert to previous' control. In the event of a problem the sysadmin can easily remove it, or if liked the sysadmin can move it into a final version, ie. install it completely.

My wich is play PS2 game on the Desktop.
But I think that is improsible because there are involve more parties.
In my dream, there should be fun & many people will like on it, such as my children.

My wish list:

* Better Hardware Detection
* Easy to use VPN to Windows VPN server
* Easy to setup Encrypted File System

I think the "not ready for purchase" list is a brilliant idea and is sorely needed.

I heartily agree that many Windows *and* Mac users are reluctant to purchase hardware that can't ever be switched to Linux. Even though applications for Linux aren't currently mature enough for my use in professional audio production, I only buy hardware that is supported on Linux. I also only buy proprietary software that supports the widest possible variety of hardware choices as well as open standards.

This should be more of a community issue wherein actual end/business users can join to add different perspectives and/or needs of an actual desktop. Is there a way to do this?

Is there any information anywhere on the net about UL? I'm interested in finding out more, but I can't find anything on google. Sounds like this project is just in the planning stages right now, but I'd love to see more.