Macbook, Software

MacBook Word Processing Part One

Since getting the MacBook I’ve been in need of a decent Mac word processer. I have NeoOffice and OpenOffice for the heavy duty stuff (and to be honest the ports are a bit heavyweight/ponderous too so you’re rarely tempted to crack them open for writing a little blog posting like this) but there is then a gap between them and TextEdit (which is what I am using now). The OpenSource alternative is AbiWord and I cannot deny that it is fast, usable and has some of the necessary functions for article writing (such as Word Count) however there is no denying that the WYSIWYG element of the processing window leaves a lot to be desired particularly in the area of font kerning. The best commercial alternative I’ve found so far is Mellel. That has the required features and a pleasing appearance as well as native performance.

However it also has its quirks such as a poor styling system, a strange desire to hyphenate words rather than do normal paragraph alignment. Hyphenating words automatically is only really of interest if you want to print the document from the application which is actually, in my view, the minority case. Most times my work is either transferred to the web or sent on to a publisher who wishes to import the raw text and styles for formatting according to their tastes. The major problem though that affects both Mellel and Abiword and really makes me hesitate to purchase a license to Mellel is lack of a “Zoom to screen size” option. The MacBook’s wide screen means you really need to able to seamlessly stretch the program to the width of the window and then have the text automatically resize to an appropriate size. There is nothing worse than looking at a long narrow column of tiny text that leaves two-thirds of your widescreen unused.

In the end the decision was made for me by Mellel, the evaluation license ran out and despite downloading a later version of the application I could not renew the evaluation. The end of the evaluation meant I was locked out of my documents, or rather I could not export the document to external formats (fair enough) and I could not cut and paste from the documents I had created during the evaluation period. This last restriction really bugged me, partly because it was shit (you could still take text from the buffers it was just more complicated to do) and secondly because if you are genuinely evaluate something and decide not to use it in the long run you should be able to transfer the content of your documents while not using any of the fancy features the application offers.

This kind of thing is the argument that is used against closed source and I for one would prefer to spend time and effort on Abiword despite its deficiencies because the effort will never be wasted and the software will never be taken away.

Standard
Java, Software

NetBeans 5.5 Beta 2 versus Eclipse 3.2 (Round One)

At last I’ve downloaded both IDEs and installed them on the MacBook and the work PC. The first thing I discover is that scrimping on the MacBook memory was a bad move as NetBeans likes a lot of it. Eclipse also likes a lot of memory but seems to do a bit better with what is available.

First test! Auto-completion: the productivity feature I use most everyday. The basic test is to have a declaration of the type Number n = new… and then hit auto-complete. What I am kind of hoping is that I get presented with a list of sub-classes of Number; Integer, Long and so on. Eclipse kindly goes for Number as the completion, which thinking about it is probably what I want to do 80% of the time. NetBeans on the other hand goes for a Burton on the MacBook, switching to the PC tells me why. The auto-completion code is trying to build a list of every possible class it knows about including XML and CORBA classes. Neither is perfect but Eclipse is closer here.

Changing to Number n = new Int… closes the gap a lot. Both go with an alphabetical search rather than looking at the context but again NetBean falls foul of checking too many possibilities and not favouring the java package over anything else. It is also noticeably slower than Eclipse on the PC (I accepted the speed on the MacBook but on a low-end development machine I would have expected comparable performance).

So, my first impression is that Eclipse is faster to produce code in. However I need to be able to leverage that, what I probably want to do is write a body of code in Eclipse and then switch it between Eclipse and NetBeans according to what is stronger in the area of development, testing and deployment I want to do later. NetBeans does seem to import Eclipse projects; Eclipse (so far) does not seem to know what to do about NetBean projects. This is a real shame, I suspect there is a plugin that does it but I haven’t found it yet.

Standard
Software

Pod Receiver People

I quite liked jPodder when I first used it but the last release or two have been hideous with all manner of issues, particularly when I installed the Java 1.6 beta

It’s handy to be au fait with Java when looking at these problems and most of the problems with it seem to come from silly attempts to not be a Java program. Like for example using the Systray4J jar which is broken. I don’t expect a Java program to integrate fully into the OS. It’s no big issue, what I wanted from jPodder was a Podcast Receiver that would allow me to use BitTorrent to share the download burden amongst all the other client users. That’s what I wanted but I’ve given up for the moment and switched to using Juice. It’s not exactly what I wanted but its lightweight and it downloads Podcasts quickly and with minimum fuss. It also doesn’t try and download a set of default podcasts unlike jPodder

Standard