For somebody who misses Lotus Improv, used Quantrix Modeller, and has evaluated XCubes, this feels about right. Unfortunately web-based solution (particularly those that cannot be self-hosted) are basically verboten to those who (such as myself) manage the kind of work that probably require this kind of instrument.
I honestly wish somebody would build something like this in the spirit of Lotus Improv for modern platforms, with modern creature comforts, as a native application.
That’s very interesting, but to be perfectly honest it’s almost the exact opposite of how I’d tend to think of things whilst building or manipulating a model. I don’t want to offend anyone, but it feels baroque, like something bolted-on, and it kind of breaks the whole “spreadsheet extruded into arbitrary dimensions” (cube) metaphor.
(Such as how in the accursed Excel, at least up to the last version I was blighted with having to use, one couldn’t take a cell of a Pivot table as the input to another formula elsewhere.)
Just watched the video. I think it is a different approach. The spreadsheets in the video seem row-oriented. It goes more in the direction of a relational database. CubeWeaver is multidimensional, so you don't need a group by function. You can group by a dimension just by creating a new worksheet with less dimensions and copying the data from the original worksheet using a formula. You can also group by an attribute using the JOIN function.
As an aside, what did you find missing from XCubes (not affiliated anyway, but I've often thought about building a similar product as Javelin or Quantrix, but sitting on top of JupyterLab).
Have you had a chance to check out Quantrix Qloud? Its basically Modeller in the browser, with reduced functionality, though the team is working to expand the feature set each release.
I don't think a multidimensional spreadsheet (like Lotus Improv) makes sense for a single user. Multidimensionality introduces a lot of complexity and it is only justified by very complex models. Apart from some special cases this complexity is only there when at least 5-10 people are entering the data. Web application is much easier to rollout in a corporate environment. And I don't see any drawbacks. I don't have MS Office for example and use only Google Sheet for a long time already.
As a single power user of such software, I assure you that multidimensionality is a characteristic of the data and not a feature that’s convenient only once one has multiple users.
The classic product/channel/geography relationship is one of those. There’s a myriad others.
Besides, Improv had many other positive features beyond multidimensionality (including the separation between data and formula).
It's usually a company-specific interpretation of software validation requirements (imposed by regulators) that software cannot be changed (ie updated) outside of the company's control. Web apps are fine as long as the company can host and perform upgrades/maintenance themselves.
Uhm, no. I assure you the powers-that-be in my firm would totally lose their marbles if highly reserved data were crossing out into some web-app hosted who-knows-where and viewable by who-knows-who.
If you choose the on-premise options no data will ever leave your company. Administrators will install the software on a server inside your company and will be responsible for updates, backups and security of the server.
A native application is a stand-alone, non-electron application whose entire stack (runtime, business logic, data engine, and information store) resides on the user’s machine.
‘Hosted’, to me at least, means that you’ve got an augmented web server somewhere on your intranet that you connect to, and that displays its interface through a browser on the user’s machine.
The two terms are mutually exclusive, in my understanding (except of course when one is speaking explicitly of “hosting a web server”, but I’m sure you catch my drift).
Egeria was just a technology demonstrator. CubeWeaver is a cloud application which can scale over multiple servers, supports user authentication, model sharing and redundant data storage. I also added a lot of features that the pilot users were missing (like data import, csv import, aggregation, various formatting options, etc...).
I would suggest to test the cloud version with some dummy data first, so that you know that the application is ok for you. On premise version is also available, but not for free. You will also need either some Linux know how or my support to install, maintain and keep your installation updated.
Congrats. Nice to see some innovation in the spreadsheet space. Often you need 3D (or more) data and excel sucks as you need to create a pivot table every time. R is good but that's coding, not always in the mood for that.
This looks tremendously useful; it's easy to imagine its adoption taking off given the obvious benefits over traditional spreadsheets. Thanks for sharing! And good luck! :)
I don’t know if it does or not, but I’m curious to know why this interests you, in the sense that if it works it doesn’t really matter to (potential) end-users such as ourselves how it is implemented. I’m not denouncing the question as illegitimate, I’m just... genuinely curious.
Thank you. It is an Angular 9 SPA frontend with a .NET Core backend and an SQL database (MySQL or SQLITE) as a data store. I use several open source libraries. Handsontable is used to display the grid for example. A complete list is available in the help->about dialog.
I'm curious, do you translate the multiple dimensions to a fully normalized relational structure, or do you also use complex structures like JSON inside the backend database?
The masterdata is stored as JSON to avoid creating/dropping columns and tables as list/columns are added and removed. But an option to have a fully normalized relational structure in an on-premise installation is on my feature list. It could be useful for enterprise clients who could use the data directly for reports and other applications.
I don't think AirTable is a competitor in any way. As far as it is something like Sharepoint. I think the logo was created before AirTable was there at all.
I honestly wish somebody would build something like this in the spirit of Lotus Improv for modern platforms, with modern creature comforts, as a native application.
Well done though, it really looks enticing.