Domain: ckeditor.com
Stories and comments across the archive that link to ckeditor.com.
Comments · 5
-
Re:Noooooooooo!
Yes, it's just an option. The monochrome theme is just one you can choose from and presumably the default as of this new version. You can choose a colour theme if you prefer, or make your own. Try the demo linked to in the summary, it uses a colour theme.
The focus of this story shouldn't be the new theme (it looks nice to me but not really a big deal) but the new functionality this version brings, mainly being the inline editing.
-
Re:CKEditor plus your favorite CMS
[......] I wold recommend WordPress (another demo) if you don't intend to develop on it.
Developing with Wordpress is at least as easy as it is with Drupal. Writing plugins is simple and the Wordpress Codex has a vast amount of good documentation.
-
CKEditor plus your favorite CMS
CKEditor plus your favorite CMS. For me that would be Drupal, but I wold recommend WordPress (another demo) if you don't intend to develop on it.
On the desktop, I recall Dream Weaver producing dirty code, though that may have changed. I wouldn't bet on that for SharePoint though. I preferred HotDog and Composer --which are still ghosting around-- before switching full time to Emacs. In short, native apps are dead, and you could as well use LibreOffice. Its tag-fu is oke'ish. Now get off ma cloud!
-
CKEditor plus your favorite CMS
CKEditor plus your favorite CMS. For me that would be Drupal, but I wold recommend WordPress (another demo) if you don't intend to develop on it.
On the desktop, I recall Dream Weaver producing dirty code, though that may have changed. I wouldn't bet on that for SharePoint though. I preferred HotDog and Composer --which are still ghosting around-- before switching full time to Emacs. In short, native apps are dead, and you could as well use LibreOffice. Its tag-fu is oke'ish. Now get off ma cloud!
-
Re:Important differenceMaybe the US is better. I'm in Hong Kong and these are what I see with local graduates:
- Theory - they can recite to you how to write recursive quick sort if you ask them, but they still write unnecessarily complicated code that runs in high polynomial times or even exponential times nevertheless. Worse, they don't even know why their program ends up running slowly afterwards. When the time comes for optimization, they take blind guesses instead of using proper profiling tools.
The most common reasons for slow programs aren't never about someone using bubble sort rather than quick sort, or performing random access on a linked list rather than a skip list - that almost never happens. If a particular algorithm is very important for a particular operation I would have told the programmers beforehand. The most most common reason for slow code is because somebody doesn't have a good sense of design and a good sense of what's happening in library functions, and did something stupid without knowing it. 4-year (it's 3 years in HK) college alone doesn't help that. In fact, what I've seen is that university graduates grossly overestimate their ability at it. Technical college people would just look at the profiler or timer outputs and optimize accordingly, or ask me if there's something they can't understand (uni. graduates hate asking, for some reason). - Exposure to lots of different programming and design styles - everybody is bad at it here, except those who go to learn it themselves. Everybody in the CS or Comp. Eng. programs need to learn Assembly (x86 and MIPS), C++, Java, Scheme and Prolog - everybody hates the Professor who teaches Scheme and Prolog (who happens to be my FYP supervisor, but I actually liked him) because they don't see how they're used in real jobs. Almost everybody hates the C++ courses because their understanding of memory management stops at new and delete, and thus the most common thing that anyone's C++ program does is segmentation fault. Only those who learn things themselves would care about object ownership and memory debuggers like Valgrind. Java is perhaps the only thing that most people can actually code in, but my university's courses are mostly taught in C++. People do well in asm projects - but that's only because those projects are exceedingly simple. A game of mastermind in MIPS asm is considered "very difficult" here.
- Database design - they can draw E-R diagrams. It stops there. What's normalization? Why do you need indexes? What's a constraint? Why are you asking me to search Google? Fuck. I just wasted a lot of time explaining exactly these things to a university graduates 2 weeks ago, when I asked him to design a relatively simple PostgreSQL db schema with very clearly laid out requirements. "What did YOU do in your database course projects, at all?!" "My course db schema only had three tables!" I gave plenty of lecture on that guy and had him redesign the schema, and he got it after 3 tries. But had I not been busy at other tasks, I could have done it much quicker.
Oh, and forget about asking these guys to use the command line client (mysql, psql, mysql-dump, etc.) - they can't navigate a Linux shell for the life of it. You HAVE to give them a GUI or web based management interface like phppgsql. - Strong English writing and speaking skills - I guess it's less of a problem in the US. Hong Kong is far better than mainland China here - you get grammatically correct English most of the time. Then, it stops there. They don't know how to use it to convey meaning. Code documentation looks like something written by a bad AI chat client - what the reader gets from the English can very often be totally different from what the programmer tried to convey.
btw, I'm a recent university graduates here and I'm also an employer here (I worked for a commercial open source project in my first two years after graduation and I'm now running my own startup) so I know both sides of the picture. Except for a few, unive
- Theory - they can recite to you how to write recursive quick sort if you ask them, but they still write unnecessarily complicated code that runs in high polynomial times or even exponential times nevertheless. Worse, they don't even know why their program ends up running slowly afterwards. When the time comes for optimization, they take blind guesses instead of using proper profiling tools.