Kexi Project – an open-source visual database applications creator | Hacker News Reader

Kexi Project – an open-source visual database applications creator

kexi-project.org

79 pointsnetworked37 comments

The following sections on the website would be useful

* roadmap : planned features with rough timeline, and an option for the public to be able to request/prioritize/vote.

* comparison : a list of features of MS Access that Kexi can do, and more importantly, what Kexi can't do yet. Its completely ok to state that something is in the roadmap, or even that it currently is not in the works.

This is very important to potential MS Access users or those who understand that world, who might consider migrating.

I don't know if there is a roadmap but you can request and vote on features in their bugzilla: https://community.kde.org/Calligra/File_a_bug_or_wish

For the other thing, I don't know if it's up to date but there's this page: http://www.kexi-project.org/wiki/wikiview/index.php@KexiComp...

Kexi uses KDE's bugzilla. There you may request features or vote for the ones you want.

https://bugs.kde.org/buglist.cgi?quicksearch=kexi%20severity...

Thanks, these are very good points.

Coincidentally, yeah, comparison and feature sheet is what we want to have soon, probably near the come back of the Windows version.

And we're looking for talented volunteers to prepare a new more modern web page. Something like https://marble.kde.org and mobile-friendly.

Priority fixes and features for Kexi can be implemented based on donations:

https://community.kde.org/Kexi/Contact#Donations

wow... didn't know this was still going. I remember reading about this (probably from slashdot) in 2004/2005...

last couple projects of mine have been rebuilding access and filemaker stuff for the web. filemaker, as nice as it is in some respects, introduces relatively high licensing costs (at least relative to my client's test pricing). i guess had he used kexi it wouldn't have been a problem... :)

Hi, Jaroslaw Staniek, Kexi maintainer here. Nice to meet you and I find the comments interesting. Feel free to ask me anything about Kexi :)

I'll try to answer some questions.

PS: came here alerted by this post :) https://www.reddit.com/r/hackernews/comments/3p8e9z

I don't -want- to be negative, but how relevant is this (* nix only, runs on single machine) in a world ruled by browsers?

My gut feeling says the Access world is dying rapidly, and if that is true, a * nix version playing catch-up does not have a bright future.

Actually, genuinely curious here, does anyone know of a solid (popular) equivalent of MS Access, but web-based? The feature I'd put emphasis on is rapid point-and-click form creation, not the database and tables part (for that of course there are plenty).

I've developped a few web applications and forms along the years, but never could I get something together as quick as I did when I decided to try MS Access, for a prototype app. E.g. I've used Django, yes I can create forms pretty quick, but with Access with _no programming_ and 2-3 hours of familiarization I could put up forms with create/edit/delete functionality in something like 30 minutes, along with the underlying schema.

I haven't touched it since I created that prototype, but I was left wondering what's the equivalent in the webapp world, for the basic CRUD use-case.

For some reason, people don't think of it in this context a lot, but this is EXACTLY what Salesforce.com does. In addition to the usual "Sales Cloud" licenses, there is a much cheaper "Force.com Platform" license that doesn't have any of the core sales features but gives you the database to play with.

100% point-and-click creation of new objects, relationships, forms, etc... Easy baked in reporting, import, and export. Workflow automation totally configurable without code. Thousands of apps on the AppExchange to add interesting features. And of course, when you need to add code, it's very easy to get started if you know OOP.

If you frequently have to create quick web-based systems that would have been MS Access, go check out https://developer.salesforce.com/trailhead, grab a free SFDC org, and build something cool.

My employer was looking for a business process integration platform. We looked at Salesforce, MS Dyanics, Oracle. We decided to go with Django + Posgres. We're in the process of converting 100's Access apps and management spreadsheets at a fraction of the cost £170 per user per month. We where looking at over £200K PA before we started talking about systems integration. Similar costs from other suppliers. The management just wasn't interested in spending that kind of cash before we had anything to show for it.

Fieldbook (my project) has a pretty solid version of tables including a form view (detail page) for each record. It doesn't have a form builder yet, but we're definitely planning on it. We have had people switch from Access with good results. Beta invite for you: https://fieldbook.com/?bc=HN0818

Mendix is one such tool (or platform as it also includes a Cloud offering for hosting the applications). It has rapid application development for web/mobile apps. At its core, it has data schema creation, form building and visual application logic modelling. This is all included in one IDE which allows for integrity checks across the board (forms, logic and database). Competitors include OutSystems, Kony, etc.

It is proprietary and mostly focused on large companies.

Full disclosure: I work for Mendix

Maybe the problem when I was searching was that actually MS Access is a narrow tool (centered around the database), whereas platforms that replaced it (RAD for mobile/web) offer a wider range of features, so searching for an "MS Access replacement" is searching with the wrong keyword given how these tools have evolved.

Still if I look at your site (Mendix) and my mindset is "I just want a few tables and a few simple forms", my guess is that if I were a non-developer I could be a bit overwhelmed. But as you say Mendix targets large companies.

Caspio seems to be quite popular. http://www.caspio.com/ And some more: https://www.knackhq.com/ http://www.ragic.com/intl/en/home https://www.zoho.com/creator/

There's Airtable, and although the form feature is rudimentary, it is point and click: https://airtable.com.

Thanks, very nice, bookmarked as usual at https://community.kde.org/Kexi/Links#Competition

Not open souce but FileMaker can be accessed by http too: http://www.filemaker.com/

I wouldn't recommend it though - as a long time FileMaker user and developer, I couldn't be happier to finally get out of this platform by learning a real web framework (Rails for me, Phoenix is on the horizon though).

FileMaker has its strengths, but the web isn't one of them. The results are subpar in my opinion and will always feel like a rapidly converted and FileMaker-branded, bloated and slow webapp. Just look at the generated HTML - it's an endless tree of nested divs.

I had a job working on Oracle Application Express.

On an awfulness scale from one to ten, it ranks as cthulu.

It's proprietary, but has a long history as an MS-Access (and other DBs) app maker : http://www.alphasoftware.com/ms-access-mobile.asp

Thanks for the pointer. I searched a bit at the time (2 years ago) and found a few platforms for rapid app development. I guess I didn't search hard enough, or I simply gave up because there was no obvious choice, like MS Access used to be. The person I developed the prototype for was also very wary of smaller platforms due to lock-in, risk of the company going bust, etc.

But Alpha Anywhere seems to have a long history and some traction, that's good to know.

You can try our Point and Click platform (http://MyDataOrganizer.com) for building custom business applications as per your workflow requirements. Out of the box team based security can be immediately used and custom security with encryption for sensitive data is also available in addition to basic CRUD.

nuBuilderPro is PHP/MySQL OpenSource alternative to MS Access: https://www.nubuilder.net

> I don't -want- to be negative, but how relevant is this (* nix only, runs on single machine) in a world ruled by browsers?

Ruled by what now? Sounds like a HN pipe dream.

Sure, I'm biased. Genuinely curious, do you know many applications being created in Access (or Kexi for that matter)? I haven't seen a new Access app in years.

Nope, not my area. But then the initial statement seemed more general.

I don't know about "dying rapidly". Access seems to be in maintenance mode since 2010 when they added web application support. It's going to be dying slowly as end users migrate to other technologies (proprietary applications and the like).

It's a bit clunky but you can run Access applications in a browser [1]. So Access is no longer synonymous with "shitty little data entry desktop application".

[1] http://www.brighthub.com/computing/windows-platform/articles...

Regarding Access-type of apps: IMHO they are as needed as spreadsheets, just less frequently. Like cars for 6 person families are needed too even if for cars for 4 are more popular.

It's not complaining but: if the percentage of people using such apps isn't increasing this is maybe also because of average level of technical education among computer users decreasing. Average user wouldn't operate at this level of specialization. Even scientists... those I know don't reject using simple spreadsheets risking numeric inaccuracies. Weak and prototype tools become the standard for them.

So simple spreadsheet apps (excel-like) that got very feature-laden over the years (statistical stuff, etc.) that clearly introduce on a concept of a strict matrix structure are very popular and they don't get replaced by web interfaces. Conversely, the web interfaces try to emulate the original UX with more or less success (Google Sheets, MS Office online).

Simplistic explanation of Access type of apps could be that they introduce data types and more on top of what's know in spreadsheets. There's always market for that but the demand needs to be realized (created?).

Ideally the database type of apps could learn some agility from the spreadsheets world. Newer MS Accesses offer implicit adding of fields in a 'spreadsheet' way in the data view and altering form/report design in the data view as well. I'd like to have that in Kexi too. Maybe even the app should welcome the user with an empty sheet of cells by default?

We have a place for discussions on the forums https://forum.kde.org/kexi and also mailing lists if someone is interested.

If it is legacy application built on Access and does not require much maintenance then it can continue running.

If it requires switching to other platform then it requires effort (1)to find new platform (2)migration to new platform and results in dying slowly.

You're probably right. I think that the number of new applications created with Access is decreasing pretty fast, but the number of active existing applications will decline slowly.

Another app (just for Linux) for throwing a few controls on a Form and making a database front end is this :

http://data-forms-action.com

Doesn't support database creation, just Form building.

It would be nice, if there was a platform like Kexi that allowed you to make not just desktop apps, but mobile apps that allowed large (>1GB) local storage of data on a phone or tablet.

Filemaker has this for iOS, but not android.

It may exist, but I haven't see it.

Right, in 2006 this was a target for quite an investment thanks to Google's SoC funds. We had a lighttpd server bundled with Kexi and it was all the way to expose Kexi projects to the web environment with a 'single click deploy' to server. Unmaintained since then, but we know how to do that properly and for the modern web.

In some way it's even better the 2006's solution isn't in use, we can start fresh. For this volunteers are needed: web developers/users/designer, not necessarily C++ hackers.

Some jquery/bootstrap components can make the data grids and forms work well.

One challenge would be maintaining the native mobile platforms. Because of local storage limits on HTML5, I think you need to use more then the stock browsers on get it done.

Yes I believe so. how do you feel about SQLite for local storage?

Kexi picked SQLite as the default backend in 2004, and it was a good bet. For native mobile there's Qt Quick for the GUI with which large part of Kexi is compatible. A database creation/handling layer has been extracted years ago and now it matures under name KDb: https://community.kde.org/KDb

I think SQLite would work well for most local storage application. It also helps that it is well known, well documented, and supported.

Reply on news.ycombinator.com