Re: [Kexi] Linux Voice review, Kexi wins in the Databases category

2015-12-14 Thread cl
Jaroslaw Staniek <stan...@kde.org> wrote:
> Hi,
> December 7: Linux Voice review, Calligra included.
> Kexi wins in the Databases category:
> 
> http://www.linuxvoice.com/group-test-office-suites
> 
Hmm, I'd hardly say it wins.  After saying that they weren't sure that
the 'desktop database' is a good idea in general they said:-

"This leaves graphical database tools with a very small target
audience. Should you find yourself a member of this exclusive group,
we’d recommend Kexi, but only marginally."

-- 
Chris Green
·

___
Kexi mailing list
Kexi@kde.org
https://mail.kde.org/mailman/listinfo/kexi


Re: [Kexi] Linux Voice review, Kexi wins in the Databases category

2015-12-14 Thread cl
Jaroslaw Staniek  wrote:
> >>
> > My main reason for not persevering with Kexi (I did try it a while
> > ago) is that it is effectively proprietary in the way it uses
> > databases.  You really can't share data between Kexi and other
> > applications.  I have, for example, a lot of sqlite3 databases created
> > by a DokuWiki plugin.  I'd love to be able to access and use them with
> > Kexi, but I can't.  I could import them into Kexi but then the
> > Dokuwiki plugin could no longer access them.
> >
> > Simply by making kexi databases have the unique .kexi suffix I believe
> > you are putting a lot of people off.  If one could at least *read* a
> > Kexi database directly it would help.
> 
> Thanks for the notes Chris.
> 
> Agree here and this is planned as a major new feature of Kexi 3, and
> has been discussed on the forum:
> https://forum.kde.org/viewtopic.php?f=221=109559
> 
> (Note, the Predicate component is now called KDb)
> 
> Features such as importing SQLite filed won't be needed then:
> https://bugs.kde.org/show_bug.cgi?id=125843
> 
Excellent.  I think separating the 'actual data' tables from the
Kexi configuration is what's necessary.  I've used other applications
which seem to manage this OK.  I suppose it's easier with a database
server as opposed to sqlite because each app can create tables on the
server and there's little risk of them disappearing whereas sqlite
files might get moved or deleted.


> 
> >>
> >> What exact features of simple table creation do you mean?
> >>
> > I want to be able to edit a table in place on a tabular form.  Table
> > creation I am quite happy to do by other means, I just want simple,
> > straightforward data entry the way that Access does it by default.
> >
> > Having created the table one is presented with a 'spreadsheet' form
> > where existing data is displayed (scrolling if required) and I can
> > edit existing entries directly and add new rows.  By default Access
> > shows a 'new entry' blank line at the bottom of the table.
> >
> > Editing entries needs to be very straightforward, just click on a
> > field and enter/delete text.  No pop-up forms, double-clicks or
> > anything.
> >
> 
> That's the way of Kexi, right? Especially the "new record" line.
> Kexi: http://i.imgur.com/O1W6l4L.png
> 
?? I went to the above URL and got what looks like a single field
entry in the middle of a blank screen.


> If I remember correctly there are no popups except one for asking
> about removing record(s) which is better handled in Kexi I think
> because there's a checkbox to dismiss the question for the future
> cases. In Access you need to use VBA or separate design mode options
> to disable the questions.
> 
> Feel free to explain what is exactly missing in the area of data
> entry/editing as this is very important part of Kexi.
> 
> The only difference to Access is that Kexi is closer to spreadsheets
> when editing a cell: you need to 2xclick or press F2 to edit existing
> value. This difference is a result of observation that navigation and
> copying/pasting of data is much easier.

I think that's one thing I don't like, I *HATE* double-click in all
its forms.  I want to just click in field and start typing, also I
expect to be able to TAB to the next field with no mouse interaction
needed at all.

-- 
Chris Green
·

___
Kexi mailing list
Kexi@kde.org
https://mail.kde.org/mailman/listinfo/kexi


Re: [Kexi] Linux Voice review, Kexi wins in the Databases category

2015-12-14 Thread Jaroslaw Staniek
On 14 December 2015 at 13:48, Swarup  wrote:
> On 12/14/2015 07:22 AM, Jaroslaw Staniek wrote:
>
> On 14 December 2015 at 13:13,   wrote:
>
> Jaroslaw Staniek  wrote:
>
> The exclusive group or RAD users is not small but severely distributed
> so quite hard to reach. Misused spreadsheets for example are
> competitors too. There's always room for IT projects without any
> reasonable budget assigned.
>
> Yes, that's my pet hate, people using spreadsheets for storing tables
> of text information.  :-)
>
> That triggered creation of this chapter of the handbook:
> https://userbase.kde.org/Kexi/Handbook/Introduction_to_Databases/Database_and_Spreadsheet
> Nothing revolutionary and can be further extended :)
>
> PS: And I am unsure why would the LV claim there's no client-server
> architecture in such RAD tools if all database engines (but SQLite)
> have always been client-server. Maybe it's about the app and GUI on
> the server side?
>
> A good, easy to use front-end (like Kexi may become or MS Access sort
> of is) would encourage more use of databases.
>
> I've been searching for a decent replacement for MS Access for ages
> and still haven't found one.
>
> My fundamental requirement is soemthing like Access' table view that
> allows very simple creation and updating of a table.  I've yet to find
> anything.
>
> Yes, sometimes I am looking into ways to make Kexi even simpler to
> use, so if user wants it - it can be in the middle between a
> spreadsheet and a db creation app. Ad-hoc creating table designs
> (while entering the data) is one of the things, quite hard to
> implement though. Interesting, after the idea came MSA (maybe 2003?)
> added similar feature.
>
> But before it's reality we probably need to get reliable alter-table
> feature to work.
>
>
> What exact features of simple table creation do you mean?
>
>
> I feel this middle ground between a spreadsheet and a db creation app is
> incredibly important and will increase the value and use of Kexi among the
> general linux user base exponentially. My estimation is the the majority of
> potential Kexi users will be wanting it for this feature. That is what a
> great number of Access users use Access for. These people are not terribly
> technically inclined, and like the table format of Access better than a
> spreadsheet format, for their purpose of storing text-based data. I am one
> such person, and as a Linux user have been using Kexi for this middle ground
> for several users-- although frankly it is far from satisfactory compared
> with Access, due to lack of ability to alter the configuration or appearance
> of a table after it has been created.

Yes, that's a #1 wish from many people
[1] https://bugs.kde.org/show_bug.cgi?id=125253

>
> So in short, I agree with you Chris, that for this middle ground between a
> spreadsheet and a db creation app, what I think you mean by the alter-table
> feature, is more important than the simple table creation. Going into data
> view to make the initial table is no big deal. But after the table is made,
> then on the fly the following features are I think critically imporant:
>
> 1. Ability to create new rows and columns later on, in table view.
> 2. Ability to rename columns later on.
> 3. Ability to hide columns and rows in table view.
> 4. Ability to move columns in table view.
> 5. Ability to alter the number of characters allowed in a cell after the
> table is made, without losing all the data.
> 6. Ability to highlight and copy the cells of limited number of rows and
> columns, i.e. a sub-section of the table.

Of course physically creating table from scratch based on data the
user entered is trivial. The points you listed are similar to the ones
in the [1] wish.
It's even known how to get there:
https://community.kde.org/Kexi/Plugins/Tables/Simple_alter_table

Just the development is more expensive than not :)

> Some of these features may have already been implemented in your later
> versions, as in Ubuntu I cannot upgrade to later versions (fixing this is
> also incredibly needed). I am running version 2.8.5, although I would love
> to be able to upgrade if there were a reasonable way to do so.

Kexi 3 will be in separate repository and won't have too many
dependencies of KDE frameworks.
I do expect this will simplify preparation of binaries for most
popular OSes like Krita has, distributed alternatively. Maybe this
include some Linux packages -- contributions in this area is more than
welcome.

The goal is that user's will no longer be forced to stay with ancient
versions -- this isn't freedom :)

-- 
regards, Jaroslaw Staniek

KDE:
: A world-wide network of software engineers, artists, writers, translators
: and facilitators committed to Free Software development - http://kde.org
Calligra Suite:
: A graphic art and office suite - http://calligra.org
Kexi:
: A visual database apps builder - http://calligra.org/kexi
Qt Certified Specialist:
: