Making KDE software the #1 choice for research and academia
Open, Needs TriagePublic

Description

Description

There is a lot of yet-untapped potential for the usage of KDE products in the research and academic sector (henceforth called R&A). This proposed goal would target all fields of research and academia.

The R&A sector has a lot to gain from using our software, for example:

  • The flexibility of our products allows them to be adapted to the often very specific needs of R&A institutions
  • They are under constant pressure to reduce cost, yet several institutions can pool the resources they have in order to improve Free software that they can all use, meaning they don't have to sacrifice quality for cost
  • Proper scientific procedure actually requires that the software used to make calculations is at least open-source in order to ensure reproducibility (and awareness of that is growing in the scientific community under the label "Open Science")
  • Much of the software they use is already Qt-based (see e.g. Kitware)

On the other hand, KDE could also benefit a lot from our software being used in that sector:

  • If students use our software on a regular basis, chances are good that they will like it and want to still use it later in the workplace
  • Students are good candidates for becoming KDE contributors, either because they get interested in hacking on the software they use, or because their university makes them work on it as part of their courses
  • As said above, if several institutions pool their resources, they can provide significant development resources and / or funding for KDE software

KDE already has a lot to offer to the R&A sector:

  • WikiToLearn is a great way for students (but also researchers) to share their knowledge
  • Kile is the best (La)TeX editor I could find (I've searched long for an equivalent application because I was not able to get it to run on my Windows machine at the university, but could not find one that even came close)
  • RKWard is by far the best R GUI / IDE I could find (same as above)
  • KDE Edu has lots of great scientific applications (e.g. Cantor, KAlgebra, Step, KStars, LabPlot, ...)
  • Plasma is perfect as a special-purpose desktop

However, our products are still unknown to the vast majority of students and academics I've met, and Plasma is certainly not the default desktop in most academic and research institutions. That should change.

I see three potential reasons (or a combination thereof) for that:

  1. We don't communicate the benefits of our software to the R&A sector enough (if at all)
  2. There are still unmet needs that people in that sector have towards our products
  3. The proprietary software sector does a lot of lobbying to tie R&A into their products (e.g. Microsoft happily throwing free licenses for their products at students)

What it will take

As KDE as a whole we could tackle all three fronts (though my hope is that if 1 and 2 are solved, rationality on the side of the institutions will defeat lobbyism).

For 1 we could:

  • Use the ties that our community already has with people in the R&A sector to establish more direct communication with them
  • Equip people in the sector who like our software with promo material and help them with advocating our software
  • Show up at academic conferences to present our products

For 2 we could:

  • Find out what unmet needs we can address by inviting people from the R&A sector to join mailing lists such as enterprise, kde-edu or specific application mailing lists, and conducting user research directly with them
  • Improve outreach in more modern communication channels (e.g. Reddit/r/kde, Facebook, Telegram)
  • Collaborate with the Document Foundation to improve LiberOffice's competitiveness with Microsoft Office
  • Continue to polish Plasma and the KDE Slimbook software experience to make switching relatively painless for new students and professors who are not technical experts
  • See if we can make use of research or teaching programs / funds or even tenders for improving our software to meet their needs
  • Better integrate WtL and the rest of our offerings
  • Make sure that our relevant applications are easy to find and install on Windows and macOS, since many researchers and students are still locked into those

For 3 we could:

  • Invite institutions that already use any of our products into the Advisory Board to make them our allies
  • Partner with the Open Science community (e.g. http://openscience.org/ or https://osf.io/ ) both for gathering requirements from them, and getting them to advocate our software

How we know we succeeded

  • The majority of people in the R&A sector are aware of KDE and our products
  • We gain institutions as partners who deploy our software for research and teaching
  • The majority of university students graduate having experience and familiarity with KDE software
  • There are success stories of R&A institutions being happy about improvements we've made to our software

Relevant links

<any links that will help people find more information and understand the goal better>

I am willing to put work into this

I am interested

  • Bhushan Shah
  • Jayaditya Gupta
  • Rishab Gupta
  • Aleix Pol
  • Sandro Andrade
  • Timotée Giet
  • Sanjiban Bairagya
  • Riccardo Iaconelli
  • Vijay Krishnavanshi
  • add your name

Related Objects

colomar created this task.Aug 29 2017, 11:27 PM
bshah updated the task description. (Show Details)Aug 30 2017, 3:58 AM
colomar updated the task description. (Show Details)Aug 30 2017, 9:23 AM
colomar updated the task description. (Show Details)Aug 31 2017, 9:44 AM
repinc added a subscriber: repinc.Sep 16 2017, 7:27 PM
colomar renamed this task from Making KDE software the #1 choice for science and academia to Making KDE software the #1 choice for research and academia.Sep 25 2017, 8:02 AM
colomar updated the task description. (Show Details)
evpokp added a subscriber: evpokp.Sep 25 2017, 10:00 AM

In my experience, one of the biggest impediments for students and professors in the humanities is dependence on Microsoft Word. They won't switch because they need perfect document fidelity when sharing, and a really obvious and user-friendly interface. And LibreOffice needs a lot of work in those departments if it's to be a drop-in replacement. It's not a KDE project, but putting some work into it would surely benefit not only us, but the whole FOSS ecosystem.

lydia updated the task description. (Show Details)Oct 2 2017, 6:07 PM
lydia updated the task description. (Show Details)
lydia added a subscriber: lydia.

Right now there is unfortunately no-one added under "I am willing to put work into this". Is that reflecting reality?

cfeck added a subscriber: cfeck.Oct 2 2017, 7:45 PM

Is this goal also include undergraduate schools or should this be a separate goal?

Is this goal also include undergraduate schools or should this be a separate goal?

If by that you mean college, then yes.
If you mean high schools, than of course many of the things we'd do would benefit those as well, but the focus should more be on college and university education.

In T6895#112448, @lydia wrote:

Right now there is unfortunately no-one added under "I am willing to put work into this". Is that reflecting reality?

It's a bit strange / sad (?): There are 8 subscribers, but nobody in the "Willing to work on this" and only Bhushan in the "Interested" section of the description.

evpokp added a comment.Oct 3 2017, 1:56 PM

I would be interested to see development on this and I could help with some evangelism, but right now I don't use a lot of KDE software for academic work apart from using Plasma as my DE and I don't expect this to change soon, as the relevant (cited) soft are lagging behind the competition, at least for my areas of interest.

cfeck added a comment.Oct 3 2017, 2:49 PM

As a math teacher, I am interested, but "willing to work on this" unfortunately has always failed because of time constraints. Additionally, my preference is in educational tools for students ages 12-16, which probably is outside the goals of this discussion.

neofytosk added a subscriber: neofytosk.EditedOct 3 2017, 7:30 PM

I wonder if KDE people working on projects related to R&A are aware of this proposal. It would be great to hear their opinion. Can we get them involved?

I think a first step would be to try this at a local level. Target a couple of organizations or departments that KDE can partner with and produce results. A Krita course was recently taught at a university, that's a great example of a first step.

meikm added a subscriber: meikm.Oct 4 2017, 9:04 AM
meikm added a comment.Oct 4 2017, 9:44 AM

i'm working on proposal 1) already (team RKWard). for instance, i was given the opportunity to do a software presentation at the 50th DGPs congress in 2016. the fact that i was given a full 90 min slot in the second largest hall can be seen as an indicator that there's actually a growing interest in software alternatives. i handed out USB sticks with a KDE live system, RKWard and various R packages, not only to invite participants to try it, but also to demonstrate that you are allowed to just give the software to all students and staff. it would already make a lot of sense to have a RKWard booth at such a conference. i'd gladly do that, so i guess you can count me in on "willing to put work into this". but without funding you can only do so much.

but what's true: we'd be totally irrelevant in these discussions if we hadn't put a lot of work into standalone installers for windows and macOS. that is an absolutely necessity, there's no way around it. there are already statistics courses now that switched from SPSS to RKWard, because they know how to explain to students how they can install it on their OS. stategically, a good amount of effort should be put into how that can be made easier for KDE software projects. if there was standalone installer for kontact, a lot of people i know would probably switch because of its superb OpenPGP integration.

I wonder if KDE people working on projects related to R&A are aware of this proposal. It would be great to hear their opinion. Can we get them involved?

Good point! I've reached out to the KDE Edu, RKward and Kile communities (though I couldn't get my mail through to the Kile list yet because they auto-reject mails from non-subscribers) explicitly, and apparently it already got meikm to comment, which is great!

I think a first step would be to try this at a local level. Target a couple of organizations or departments that KDE can partner with and produce results. A Krita course was recently taught at a university, that's a great example of a first step.

Interesting idea!

In T6895#112766, @meikm wrote:

i'm working on proposal 1) already (team RKWard). for instance, i was given the opportunity to do a software presentation at the 50th DGPs congress in 2016. the fact that i was given a full 90 min slot in the second largest hall can be seen as an indicator that there's actually a growing interest in software alternatives. i handed out USB sticks with a KDE live system, RKWard and various R packages, not only to invite participants to try it, but also to demonstrate that you are allowed to just give the software to all students and staff. it would already make a lot of sense to have a RKWard booth at such a conference. i'd gladly do that, so i guess you can count me in on "willing to put work into this". but without funding you can only do so much.

Good do hear that, that's certainly something we can build upon!

but what's true: we'd be totally irrelevant in these discussions if we hadn't put a lot of work into standalone installers for windows and macOS. that is an absolutely necessity, there's no way around it. there are already statistics courses now that switched from SPSS to RKWard, because they know how to explain to students how they can install it on their OS. stategically, a good amount of effort should be put into how that can be made easier for KDE software projects. if there was standalone installer for kontact, a lot of people i know would probably switch because of its superb OpenPGP integration.

Good point, being easily available on macOS and especially Windows is indeed key for establishing our apps in new sectors.
From your experience, how important is macOS compared to Windows?

I would be interested to see development on this and I could help with some evangelism, but right now I don't use a lot of KDE software for academic work apart from using Plasma as my DE and I don't expect this to change soon, as the relevant (cited) soft are lagging behind the competition, at least for my areas of interest.

This goal is about both evangelism _and_ improving the software itself, and both go hand in hand, of course: The more people use the applications, the more likely some may contribute to them and improve them. The more we improve the apps, the more likely they are to get used.

In my experience, macOS is heavily over-represented in academia here in the USA. Much less so in Europe.

In my experience, macOS is heavily over-represented in academia here in the USA. Much less so in Europe.

Okay. So both macOS and Windows it is, then. I've edited the description.

colomar updated the task description. (Show Details)Oct 4 2017, 10:52 PM
asemke added a subscriber: asemke.Oct 5 2017, 10:00 AM
meikm added a comment.Oct 5 2017, 1:01 PM

Good point, being easily available on macOS and especially Windows is indeed key for establishing our apps in new sectors.
From your experience, how important is macOS compared to Windows?

adding to what @ngraham said, macOS might not be as common as windows globally, but quite prevalent especially in academia. since this is our target here, it really is a must-have support area.

good news is that there is a KDE-on-mac community, and many applications can be built and used already. in fact, the script that we use to build installer bundles of RKWard for macOS uses the MacPorts ecosystem that was already in place.

asemke added a comment.Oct 5 2017, 3:06 PM
In T6895#112448, @lydia wrote:

Right now there is unfortunately no-one added under "I am willing to put work into this". Is that reflecting reality?

It's a bit strange / sad (?): There are 8 subscribers, but nobody in the "Willing to work on this" and only Bhushan in the "Interested" section of the description.

The description of this proposal is IMHO a bit vague for the developers in a sense that it sounds like all our projects are already "done" and cover the biggest part of the features relevant for their specifis areas and we only need to heavily invest now in promotion and publicity work. This is not the case. I'd agree here with what @evpokp said. So, I'm missing in this project proposal the part where we want to greatly improve the software we already have. With improve I don't mean here small fixes here and there but rather substantial feature improvements in order to be able to compete with other commercial and free software used in the different research areas.

I'm interested in this project and I invest already a lot of time into LabPlot right now. As mentioned above, a lot of stuff still needs to be implemented. I'm not sure I'll find some time for doing promotional work while still heavily developing in the code. This is the reason for me why I'm hesitating to add here my name under "I'm willing to put work into this". If you mention in this project also some development effort like "Add the killer feature A to the KDE application B" or "investigate possible collaborations between RKWard, LabPlot and Cantor", etc., I'm part of it without any further hesitations.

The description of this proposal is IMHO a bit vague for the developers in a sense that it sounds like all our projects are already "done" and cover the biggest part of the features relevant for their specifis areas and we only need to heavily invest now in promotion and publicity work. This is not the case. I'd agree here with what @evpokp said. So, I'm missing in this project proposal the part where we want to greatly improve the software we already have. With improve I don't mean here small fixes here and there but rather substantial feature improvements in order to be able to compete with other commercial and free software used in the different research areas.

It puzzles me that people perceive the proposal that way, to be honest: Reason 2 ("There are still unmet needs that people in that sector have towards our products") fully recognizes that there is stuff to do to improve our products, which is why all the points under "For 2 we could" are purely about improvements. Only Reason 1 and 3 are about promotion and advocacy.

The reason why I have not gone more into detail about what we should work on specifically is simply that I do not know yet what the unmet needs are. That's why I proposed to first actually ask people in research and academia what they need, and then work on providing that.
If any of you already have a list of things that need to be done for our products to meet the needs of the R&A sector, please feel free to edit the proposal and add those specific points, I have absolutely nothing against that.

So, @asemke I'm happy that you're interested in this proposal, and if you already know what specifically needs to be improved, please add it to the description. In addition to what you already know, specific imrpovement points would definitely come out of the user research we'd conduct with scientists, teachers and students.

I agree that we first need more insight what R&A really wants to define the goals more precisely. I can easily put my name under "willing to work on this" because i already work on LabPlot and plan to continue it anyway and my job is in R&A.

I work in research and teaching at a university in the area of physics, so i can give some more input here. Researchers are already using open-source software where they can. I'm thinking of
Python (with many packages), R, Octave, gnuplot, Sagemath, GCC, etc. But in some areas (especially when trained before in classes), also commercial software is used. Common names are Matlab, Origin, Mathematica, Labview, Comsol and IDL. The feature sets of these packages are almost always much bigger than compared to open-source alternatives. But sometimes "good enough" open-source software is all we need. I think there are already good websites listing alternatives to commercial packages for different fields.

Another problem is that we need to communicate a lot with administration and management, so we are sometimes forced to use certain Windows-only software like MS Office, because they do.
I don't see a short-time solution for this right now. And yes, Macs are also very popular here.

Let me know if i can be of any help.

@meikm @asemke @sgerlach @evpokp It would be great if you could edit the task description and add your name in the "I am willing to put work into this" section, just to have the list of people.

Just to clarify again: This is about both advocacy and improving our software to meet the needs of the sector, and it's a goal for the next 3-4 years, so lots of time to actually make it happen.

sgerlach updated the task description. (Show Details)Oct 5 2017, 4:35 PM
asemke updated the task description. (Show Details)Oct 5 2017, 5:10 PM
meikm updated the task description. (Show Details)Oct 5 2017, 5:17 PM
evpokp updated the task description. (Show Details)Oct 5 2017, 8:48 PM
GreatEmerald updated the task description. (Show Details)Oct 5 2017, 9:44 PM
GreatEmerald added a subscriber: GreatEmerald.

So far it looks like this goal does a good job of targeting the STEM fields. But I don't see as much attention being paid to the needs of fields in the humanities and social sciences. Folks in these fields aren't as technical, which is a blessing and a curse: they don't need as many esoteric pieces of software (Kile, RKWard, Cantor, KAlgebra, Step, KStars, LabPlot, etc), but they are more demanding of user-friendliness in the basic software and in greater need of a high-quality interoperable office suite (particularly the word processor). This is where Apple currently has a disproportionate market penetration owing to their strong focus on user-friendliness, an integrated hardware/software package, and compatibility with Microsoft Office.

If we can broaden the scope to include some of those use cases, with the knowledge that this will pay dividends for academic users in non-STEM fields, then I would be willing to climb aboard and put work into this.

That's a nice idea. I'm involved with a course that introduces Linux and programming concepts to students and I'd love to see more KDE representation. Last year we used XFCE, this year we'll most likely be using GNOME Flashback.

That said: RKWard really needs more work porting to KF5! And a few more quality-of-life improvements (the package installer is not very intuitive). But overall it's great, really useful for beginners too. That said, mostly it's competing with RStudio, which has a bunch of features that have little to do with R (debugging, knitting, Shiny package creation) but that people tend to expect these days. It's a whole lot more stable than RStudio, though, which is why I always recommend using it instead.

Kile is pretty amazing (especially coupled with Okular), and its KF5 port is great too. An easy way to install it on Windows would help (right now all the PCs in my university are Windows; it's changing to allow more choice, but slowly).

The next thing that needs improvements for my field (GIS) would be Marble. It's OK right now, but it doesn't quite match Google Earth (KML files with textures not displayed, etc.).

There are some people here with macOS laptops, but not that many. Windows dominates. I usually suggest installing Linux on an external HDD if one wants to get any real job done right now.

I'm listing myself as "willing to put work" but more on the evangelical side, I suppose.

meikm added a comment.Oct 6 2017, 8:22 AM

RKWard really needs more work porting to KF5!

FWIW, there's some ubuntu PPAs with daily builds of the KF5 port. i'm using one for work for quite some time now. if you have something particular on your mind, we're always happy for suggestions. we know that a stable release is overdue (the latest stable 0.6.5 still uses KDE 4). please help us find issues that still need fixing. i guess we should keep it about broader strategy here, and discuss particular technical issues someplace else, right?

That said, mostly it's competing with RStudio, which has a bunch of features that have little to do with R (debugging, knitting, Shiny package creation) but that people tend to expect these days.

i agree. however, i think the biggest advantage RStudio has over RKWard is a company with payed developers behind it, and a budget for marketing. they're not only producing a high quality IDE, but also have the means to do the proper advertising. most new R users these days don't learn anything else, they don't chose. i'm convinced we have some good arguments on our side. i.e., RKWard is not only an IDE but also a GUI with great features for teaching. the teachers who are using it in their courses do so because they've learned that it is much easier to teach students R with it. meaning, there's not only features in RStudio that RKWard is missing, it's also the other way around, and that is because of slightly different scopes of both tools. we also might add support for knitr at some point, but i don't see that coming in the near future, because we're actually stretched a bit thin on developers ATM. and we would't so much do it to compete with RStudio's features, but to re-implement the current method of printing results. we're discussing this for a long time now. we're happy for input here as well!

I agree with ngraham it would be nice if KDE somehow can address to those academics in the humanities and social science institutes. Word processor, slide presentations, and a reference manager (e.g. Zotero) are the main tools used here. As many her know right now MS office provides the common tools for that purpose, even when the interfaces are not friendly. It is a matter of custom and market pressure. I think in this regard KDE could collaborate with the document foundation to strengthen the integration of Libre Office and the desktop, along with what it is mentioned as campaigning the use of KDE at academic conferences in the humanities and social sciences. University and city libraries is another place to go bc usually fresh-students get their introduction to those tools in those places.

drnn1076 updated the task description. (Show Details)Oct 6 2017, 9:00 AM
evpokp added a comment.Oct 6 2017, 9:46 AM

I think in this regard KDE could collaborate with the document foundation to strengthen the integration of Libre Office and the desktop,

Yes for that. From what I see, the LibreOffice/Zotero combo already fits the needs of the HSS (though it is not without a considerable margin of improvement) and I don't think that trying to develop KDE alternative for those would be productive. But the integration with the desktop is awkward and it is hard to tear researchers away from Microsoft Office/Whatever the Apple alternative is, though the cost argument might win us the administrators.

If we can broaden the scope to include some of those use cases, with the knowledge that this will pay dividends for academic users in non-STEM fields, then I would be willing to climb aboard and put work into this.

I've focused on the STEM fields mostly because that's my background, I never meant to exclude any other fields. This proposal is a collaborative effort, so please, those of you who have more experience with what other fields need, feel free to add your suggestions to the description!

As long as it doesn't turn into a generic "Let's make our software better for people" but keeps its focus on what research and academic fields need, I'd be happy to see the proposal expanded!

ngraham updated the task description. (Show Details)Oct 6 2017, 3:01 PM
ngraham updated the task description. (Show Details)
GreatEmerald added a comment.EditedOct 6 2017, 4:10 PM
In T6895#112961, @meikm wrote:

FWIW, there's some ubuntu PPAs with daily builds of the KF5 port. i'm using one for work for quite some time now. if you have something particular on your mind, we're always happy for suggestions. we know that a stable release is overdue (the latest stable 0.6.5 still uses KDE 4). please help us find issues that still need fixing. i guess we should keep it about broader strategy here, and discuss particular technical issues someplace else, right?

Yea, I'm mostly referring to a stable rease, although in openSUSE there is no daily KF5 build (even in the KDE:Unstable repos that ought to have these sorts of things).

i agree. however, i think the biggest advantage RStudio has over RKWard is a company with payed developers behind it, and a budget for marketing. they're not only producing a high quality IDE, but also have the means to do the proper advertising. most new R users these days don't learn anything else, they don't chose. i'm convinced we have some good arguments on our side. i.e., RKWard is not only an IDE but also a GUI with great features for teaching. the teachers who are using it in their courses do so because they've learned that it is much easier to teach students R with it. meaning, there's not only features in RStudio that RKWard is missing, it's also the other way around, and that is because of slightly different scopes of both tools. we also might add support for knitr at some point, but i don't see that coming in the near future, because we're actually stretched a bit thin on developers ATM. and we would't so much do it to compete with RStudio's features, but to re-implement the current method of printing results. we're discussing this for a long time now. we're happy for input here as well!

Yea, that's understandable, of course. Indeed RKWard is great for beginners. But to give a bit of a background of the current situation here: RStudio is installed by default in the Windows PCs in the university, and the course I'm talking about covers RMarkdown and git, and encourages trying out Shiny. Last year I changed the material so that students would learn to use Git GUI rather than the client in RStudio (and still got questions from students about why not to use the one in RStudio), so that's an improvement. But right now to be able to use RKWard instead of RStudio, we'd need to basically duplicate a bunch of material, with one part explaining how to do things in RStudio and the other part explaining the same in RKWard, since RStudio has RMarkdown integration that we need to cover. That said, from discussions so far I think we're going to do that duplication work this year anyway, simply because RStudio proved to be quite unstable last year and would keep crashing (whereas RKWard is rock solid). But being able to use RKWard exclusively would really help cut down on the maintenance work.

filipesaraiva added a comment.EditedOct 7 2017, 9:26 AM

I like the idea of improve the penetration of KDE software in research and academia. Maybe it is not suitable as an effort from all community, but see it as an effort from a specific subgroup is interesting.

In past I gave some presentations about KDE software in academia, like this Okular, KBibTex, and Kile: Writing Science with KDE (sorry, only in Portuguese).

Reading the thread, KBibTex was not cited. It is an interesting bibtex manage software, in KDE4 times there was an integration with Kile. KBibTex was ported to Qt5/KF5.

Beside the scientific-oriented software, we could list some general software with great application in science. I am talking about Okular because the feature of write/read annotations in the PDF is unique in free software world; Kate, to write scripts in several programming languages with syntax highlighting and code completion; and Konsole.

jgupta updated the task description. (Show Details)Oct 7 2017, 11:28 AM

CCing this thread to @thomasfischer, the KBibTex author.

Reading the thread, KBibTex was not cited. It is an interesting bibtex manage software, in KDE4 times there was an integration with Kile. KBibTex was ported to Qt5/KF5.

Yea, I use KBibTeX as a bridge to Zotero, it's quite useful in that sense. Though the rest of it... I feel that it needs work or reimagining. Zotero is real good at getting metadata from other sources, so I hardly ever see the need to edit entries (and when I do, I edit the entry in Zotero). Even when using standalone, editing entries is easier to do by hand editing text rather than the unwieldy window and menu system of KBibTeX. I feel that it should either be streamlined (for instance, show only existing fields of the selected entry and a single "add" button) or be outright replaced by a simpler tool.

apol updated the task description. (Show Details)Oct 7 2017, 3:06 PM
apol updated the task description. (Show Details)

Beside the scientific-oriented software, we could list some general software with great application in science. I am talking about Okular because the feature of write/read annotations in the PDF is unique in free software world;

Actually Xournal also has PDF annotation features. But there's a broader point: as a former Mac user, let me bring up the unfortunate truth that Okular can't hold a candle to Apple's Preview, *especially* in the annotations department. That's the standard Okular needs to aspire to. Right now Okular's annotations have a major problem: they aren't consistently compatible with viewers that aren't Okular. Annotations are often not visible when printing or viewing with another problem. This is of course tracked by many bugs:

Anyone who uses Apple Preview's annotation features and switches to Okular is bound to be disappointed at the moment. And that's a general theme across all of KDE: some of our key programs lag behind their proprietary counterparts. IMHO addressing that is far more important than outreach and advertising. If the product doesn't practically sell itself, you're fighting an uphill battle.

meikm added a comment.Oct 8 2017, 12:11 AM

Right now Okular's annotations have a major problem: they aren't consistently compatible with viewers that aren't Okular. Annotations are often not visible when printing or viewing with another problem.

does apple's preview implement an open standard when it comes to annotations? if there is one in the PDF specs (i'm not an expert on this) it should be implemented. if apple uses its own proprietary standard, it would be a waste of time, because apple often changes those and leaves you stranded (like the multiple times they broke OpenPGP support in mail, and still do from time to time).

does apple's preview implement an open standard when it comes to annotations? if there is one in the PDF specs (i'm not an expert on this) it should be implemented. if apple uses its own proprietary standard, it would be a waste of time, because apple often changes those and leaves you stranded (like the multiple times they broke OpenPGP support in mail, and still do from time to time).

I don't know the technical details. All I know is that I can't remember the last time in the past 16 years of being a (prior) full-time macOS user when a PDF I annotated didn't print properly or display properly in Acrobat Reader, Foxit Reader, or anything else I happened to use to open it. Regardless, if we want annotations to be useful, we don't have the luxury of only making sure they render properly in Okular. Like it or not, Most people out there are using Acrobat Reader or Preview.

This is especially true of university administrators, to bring it back to the original goal. If a student or professor fills out a PDF form or annotates a PDF document in Okular and then sends it to another student, professor, or administrator who's using a Windows PC with Acrobat Reader or a Mac with Preview, and the forms or annotations don't render properly, then they're not going to blame the unimplementability or non-standard-compliance of the PDF file format; they're going to blame Okular or KDE.

meikm added a comment.Oct 8 2017, 12:33 AM

I don't know the technical details. All I know is that I can't remember the last time in the past 16 years of being a (prior) full-time macOS user when a PDF I annotated didn't print properly or display properly in Acrobat Reader, Foxit Reader, or anything else I happened to use to open it.

if it does work in both directions, it's hopefully part of the PDF specs.

Regardless, if we want annotations to be useful, we don't have the luxury of only making sure they render properly in Okular. Like it or not, Most people out there are using Acrobat Reader or Preview.

i agree with you and i'm all for full compatibility. i'm just saying it's sometimes not a lack of motivation, but closed standards or patents that keep you out (and willingly so).

Right now Okular's annotations have a major problem: they aren't consistently compatible with viewers that aren't Okular.

I researched about that some years ago (in Portuguese, sorry) and I found annotations made in Okular are very compatible with several software. Several people report me if a software could open the annotations correctly. I don't know if someone tested with some software in Mac OS, but the problem you are commenting looks like more related with the software you are using.

In my tests and my daily work I make annotations in Okular and my fellows open and read the annotations in Acrobat, Foxit, Evince, and Utopia Documents. When they reply me, I can see the annotations made by them. It is common for me make annotations in Okular and send the PDF to my Android device and read the PDF and annotations in EzPDF. Of course, when I make an annotation in EzPDF I can open that in Okular.

In the post linked above there is a PDF file with annotations generated by Okular. Download that and let's see if the annotations will be presented by the software you are using. :)

Actually Xournal also has PDF annotation features. But there's a broader point: as a former Mac user, let me bring up the unfortunate truth that Okular can't hold a candle to Apple's Preview, *especially* in the annotations department. That's the standard Okular needs to aspire to. Right now Okular's annotations have a major problem: they aren't consistently compatible with viewers that aren't Okular. Annotations are often not visible when printing or viewing with another problem. This is of course tracked by many bugs:

Half of those bugs date back to before saving annotations was even implemented in Okular. The other two are edge-cases; the Chinese character one is clearly a simple bug and the one you reported is an edge-case (and no, you *really* shouldn't "sign" documents by using custom stamps!) that is being looked into. I don't think the annotation handling is quite as poor as you make it out to be.

Usually I fill in very simple PDF's forms and annotate pdf with okular. I also exchange them with colleges and secretaries at the institute where I work. They work mostly with Win and to a lesser extend with mac. So far I have had very few problems in terms of rendering the information or the annotations, really very few. However, it might be due to the kind of very simple forms and annotations I'm processing in Okular.

Having a larger and representative swat of examples with larger and very complex annotated pdf documents in both directions, (a) acrobat and preview, and (b) okular might help spot the problems @ngraham is describing.

I'm happy that there is a lot of discussion happening here, but maybe the discussion about annotations in Okular is getting a bit too much into detail for this point in the process.
So let's maybe summarize it for now as "There are potentially some issues with compatibility of Okular annotations, so if this goal gets selected, this could be one of the things we look into, together with academic / research users.

Agreed, I didn't mean to start a whole new branch of the conversation. Just mention that it might be beneficial to also to focus on basic productivity software as a part of this goal.

I also agree with @colomar
In any case, just to keep in mind the current official status of document format handlers in okular:
https://okular.kde.org/formats.php

ruphy updated the task description. (Show Details)Oct 20 2017, 12:10 PM
ruphy added a subscriber: ruphy.
edarfoc added a subscriber: edarfoc.Sat, Nov 4, 3:13 PM
edarfoc updated the task description. (Show Details)Sat, Nov 4, 3:17 PM

I'm a mathematician working in academia, hence a heavy latex user. I've used linux for way over 10 years, with the desktop environment provided by KDE most of the time. I've used and recommended Kile, kbibtex, etc. Unfortunately I should say that, despite being an excellent piece of software, Kile has nowadays been overtaken by TeXstudio, and tools like kbibtex have been left behind in favor of more complete ones like Zotero or Mendeley. I'd love to be able to make something to revert this, hence I've written my name in the list.

maxrd2 added a subscriber: maxrd2.Mon, Nov 6, 2:25 PM

Thanks everyone for helping draft this proposal. The voting has started. If you are an active KDE contributor and have not received an invitation to the vote please send me an email to lydia@kde.org.