Over the past few months I have read several articles like this:
http://www.h-online.com/open/features/LXDE-and-Xfce-the-other-desktops-13925...
I'm not sure Xfce is a cholesterol free desktop or that LXDE is ready for prime time with non geek users. Yet I am sure that Trinity is receiving little comparative press coverage.
I'm not calling for a marketing campaign, but I would like to request all team members focus on making R14 the best release yet. To me that means three goals:
* Eliminating many, many paper cut bugs.
* All packages build with minimal fuss.
* Provide an improved web site.
The paper cut project never got off the ground. Paper cut bugs discourage non geek users. They don't care or want to understand the nuances of programming or why something fails to function as intended. Paper cut bugs are a public relations nightmare. People vote with their feet. :(
Struggling to build packages is a good way to ensure little exposure because the desktop then is not available to users. A significant portion of the discussion in this list is build issues. Simple things like autotools looking for Qt4 rather than Qt3 are frustrating to non geeks. Remember that until Trinity is provided as a regular prebuilt package option in most distros, end-users are left to build the packages on their own. That means the wiki needs attention too and should be revised for non geeks.
We have been discussing web site changes and a possible RSS feed. Those plans likely will fall into place shortly. :)
Despite what reviewers declare, I don't see Trinity "competing" with KDE4 --- or GNOME. I see the "competition" as Xfce and LXDE. I use the word "competition" in a friendly, comparative way. Long-term we should be asking ourselves questions such as whether there is anything those desktops do better than Trinity? Is there anything those desktops do that Trinity does not? Trinity does not have to be at the top of all comparison lists, but should look favorable all around.
There is one thing those two desktop environments do better than Trinity: start and exit faster. I don't know whether that observation merits an enhancement request. While I agree too much discussion is wasted on the topics, I'd like to think both can be improved in Trinity.
In the article I linked the writer mentions that "One developer in China has ported LXDE to a device with 128 RAM, 400 MHZ CPU." How does Trinity perform with such hardware? I have PI and PII machines sitting here with 256 MB of RAM. I have run 3.5.10 on those machines for several years and 3.5.10 is, well, sluggish. I suspect my problem is the video cards in my machines and I haven't fully tested yet. Yet I am guessing Trinity will fare only a tad better than 3.5.10.
The author also quoted somebody as saying "if Windows 98 and XP work quite well on old machines, why does my Linux desktop need a 1.0 GHz CPU + 1GB RAM?" I agree. I think Trinity can be a significant player in keeping old hardware running. If LXDE is going to be touted as ideal for that kind of hardware, then is that a topic for discussion for Trinity too?
One thing Trinity does well is conform to the traditional desktop model that most users are accustomed. That means users familiar with Windows. I never have had a problem with Linux desktops continuing that model and to me, the original 1969 desktop model derived in the PARC labs and copied by Microsoft and Apple developers still works wonderfully well today.
Not to mention that Trinity is far more configurable than Xfce and LXDE.
There are several distros focusing on being lightweight or fast. The simple message of that focus is many users do not like bloat but they also do not want to deal with the "crippled" world of window managers only. In the end, will distro maintainers offer Trinity as an option because Trinity satisfies such goals? If Trinity is easy to build, will Trinity be offered as an alternate desktop in major distros too?
Opportunity is waiting. Let's strive to make R14 a desktop reviewers want to review and praise. :)
Darrell
On 19 December 2011 16:50, Darrell Anderson humanreadable@yahoo.com wrote:
Over the past few months I have read several articles like this:
http://www.h-online.com/open/features/LXDE-and-Xfce-the-other-desktops-13925...
I'm not sure Xfce is a cholesterol free desktop or that LXDE is ready for prime time with non geek users. Yet I am sure that Trinity is receiving little comparative press coverage.
Time, time time. I have noticed a serious uptick of Trinity mentions in every reddit thread about KDE, GNOME3, or the whole current desktop environment disaster.
I'm not calling for a marketing campaign, but I would like to request all team members focus on making R14 the best release yet. To me that means three goals:
I think we need a slight marketing campaign and have a few people up my sleeve who are already helping develop ideas for a campaign of sorts.
Eliminating many, many paper cut bugs.
All packages build with minimal fuss.
Provide an improved web site.
Yes +3
The paper cut project never got off the ground. Paper cut bugs discourage
non geek users. They don't care or want to understand the nuances of programming or why something fails to function as intended. Paper cut bugs are a public relations nightmare. People vote with their feet. :(
KDE4.0 had this exact problem. Overall it worked well but there were many things users hit and missed, then never returned. The current stable branch is much more stable and performs without hesitation recently, but they are still paying the price in skeptical users.
Struggling to build packages is a good way to ensure little exposure because the desktop then is not available to users. A significant portion of the discussion in this list is build issues. Simple things like autotools looking for Qt4 rather than Qt3 are frustrating to non geeks. Remember that until Trinity is provided as a regular prebuilt package option in most distros, end-users are left to build the packages on their own. That means the wiki needs attention too and should be revised for non geeks.
Obviously there is the ongoing CMake effort. Though we should focus more attention on getting packages out the door, so that users won't have to build packages for widely used distrobutions.
We have been discussing web site changes and a possible RSS feed. Those plans likely will fall into place shortly. :)
Despite what reviewers declare, I don't see Trinity "competing" with KDE4 --- or GNOME. I see the "competition" as Xfce and LXDE. I use the word "competition" in a friendly, comparative way. Long-term we should be asking ourselves questions such as whether there is anything those desktops do better than Trinity? Is there anything those desktops do that Trinity does not? Trinity does not have to be at the top of all comparison lists, but should look favorable all around.
LXDE and XFCE fall miles behind in functionality and integration. I laugh at the fact they even call LXDE an environment. All components are so modularized (which isnt bad) that they are all independent. They don't even have their own window manager and use Open Box instead. Basically as I see it, the functionality provided by LXDE is that of me running Openbox, with a panel and desktop, using a File Manager of my choice.
XFCE is more of a competition, but we still beat them by miles with Konqueror and our other integration. That said, I love XFCE and it is my favorite GTK desktop.
There is one thing those two desktop environments do better than Trinity: start and exit faster. I don't know whether that observation merits an enhancement request. While I agree too much discussion is wasted on the topics, I'd like to think both can be improved in Trinity.
In the article I linked the writer mentions that "One developer in China has ported LXDE to a device with 128 RAM, 400 MHZ CPU." How does Trinity perform with such hardware? I have PI and PII machines sitting here with 256 MB of RAM. I have run 3.5.10 on those machines for several years and 3.5.10 is, well, sluggish. I suspect my problem is the video cards in my machines and I haven't fully tested yet. Yet I am guessing Trinity will fare only a tad better than 3.5.10.
What if we created some sort of "lo-fat" settings to help it run minimally and fast?
Opportunity is waiting. Let's strive to make R14 a desktop reviewers want to review and praise. :)
hear, hear!
Calvin Morrison
What if we created some sort of "lo-fat" settings to help it run minimally and fast?
We already have the basis for such a tool: KPersonalizer. Although I think KPersonalizer falls short in creating a minimal desktop.
If the section in KPersonalizer that sets the desktop settings was a KPart then we could add those settings directly into Control Center.
In other words, we some nominal work we could do what you seek. Does this sound like an enhancement request? :)
Darrell
LXDE and XFCE fall miles behind in functionality and integration. I laugh at the fact they even call LXDE an environment. All components are so modularized (which isnt bad) that they are all independent. They don't even have their own window manager and use Open Box instead. Basically as I see it, the functionality provided by LXDE is that of me running Openbox, with a panel and desktop, using a File Manager of my choice.
I just came across this article about razor-qt:
http://www.webupd8.org/2011/12/razor-qt-new-lightweight-desktop.html
Seems the desktop is dependent upon Qt rather than kdelibs. If some want to explore making TDE apps available as stand-alone, perhaps a few tricks could be learned by how these people wrote razor-qt.
Even if LXDE is more a hodge-podge of apps and utilities rather than a true cohesive desktop, many users nonetheless will perceive that environment as a competitive "desktop" to Trinity. Seems razor-qt could be thrown loosely into the discussion about "competition." Granted razor-qt is just a desktop and one without a native window manager, and offers almost no apps or utilities (yet), but probably one we should watch for reviews, features, and ideas.
Darrell
-
Calvin Morrison -
Darrell Anderson