I finally got some time to implement a small amount of my meta_tree project.
meta_tree (always lowercase) is an NT-only implementation of my own file-meta-data managing-software.
It gives you the possibility to add meta-data ("tags") to your files.
It is aiming for people who are organizing their data heavily on folders and find it difficult to find and reorganize their structure to changing needs.
meta_tree will allow you to use tags on their files and search and reorganize their files depending on this tags.
Showing posts with label usability. Show all posts
Showing posts with label usability. Show all posts
Sunday, December 23, 2007
Monday, November 26, 2007
why subversion really rocks
Subversion leaps giantly forward. Not only some opensource people are using it, now a lot of large companies try to make the switch. And I do not speak of some old fashioned CVS-users how just try new things: There are companies outside which will abandon their commercial version management systems for subversion.
I think the main reason for this is subversions model of representing version control:
At the end users are most comfortable with the most common computer metaphor:
the file(-system)
Nearly all users now how to copy, rename move and delete files. I think all developers know.
Subversion breaks not with this metaphor, it just enhances it with a new dimension: time. This makes the learning curve so shallow. You do not need anything to now about branches, tags or anything else. Just this: a commit saves your versions into the repository and creates a new revision an update will get changes from your coworkers into your workingcopy.
Of course there are voices against this simple model: some people think it is a problem that svn doesn't support native tags or branches but I think as long as you can map this usecases on your filesystem more people will understand what is happening. So if your branches/tags are just plain directories, it is much easier to grasp than an abstract concept for people who do not work day to day with version control.
I think the main reason for this is subversions model of representing version control:
At the end users are most comfortable with the most common computer metaphor:
the file(-system)
Nearly all users now how to copy, rename move and delete files. I think all developers know.
Subversion breaks not with this metaphor, it just enhances it with a new dimension: time. This makes the learning curve so shallow. You do not need anything to now about branches, tags or anything else. Just this: a commit saves your versions into the repository and creates a new revision an update will get changes from your coworkers into your workingcopy.
Of course there are voices against this simple model: some people think it is a problem that svn doesn't support native tags or branches but I think as long as you can map this usecases on your filesystem more people will understand what is happening. So if your branches/tags are just plain directories, it is much easier to grasp than an abstract concept for people who do not work day to day with version control.
Thursday, October 25, 2007
Is this usability at its best?

I just got the Microsoft desktop indexing Engine installed via online update. I wanted to configure it, so it will not search and index every file on my computer, but only a few selected datatypes. But the dialog is a mess! You have a miles long list for each file-suffix and you cannot, I repeat: you CAN NOT switch all checkboxes on or off with one click, but you have to click on each checkbox. Please note this small scrollbar which indicates the ridiculous number of entries.
What are this Microsofties doing in their offices?
Maybe they need to click on all this checkboxes..
Subscribe to:
Posts (Atom)