TextPad or UltraEdit?

[skaemper]

Why don't we build great software?

• Nebulous specifications.
• Mutable specifications.
• Limited budgets.
• Design by committee.
• Limited (absurd) time-frames.
• We don't really understand the problem domain.

The last item is key to understanding the success of Textpad. Programmers have a very clear and complete idea of what constitutes a good programmer's editor. We understand what is frippery and what is crucial, and we understand how the software will be used. This is what underpins the success of Linux as a server OS and is also the reason that it is rather less stellar as an OS for end users.

While I agree on some of your points, I think other points are not as "valid".
If you, as a developer, are in contact with the customers/useres of the to-be-developed software, you can cope with 'Mutable specs'. May be it's not easy, may be to don't like it - but you can.
You can cope with neboulus and mutable specs, if you have a thight contact to the end users of the software.
If you have achieved this, you'll have a good chance to understand the problem domain - by talking to people who do so: Your customers and/or users.

BTW, as I'm currently looking for a new job, I think it's really stragne that so many companies more or less strictly insulte the developers from the customers. Sometimes I was told that this would provide the 'loose coupling' known from oo programming. ?!?
So loosely that one side wonders why the other side produces hard to use products.

A Limited budget is just one of the 'hard' constraints. Others are: Available time, product quality, number of features. The problem is (IMHO) that way too often developers (or development teams) are forced to work under constraints that can#t possibly work out. It's impossible to set all of these (as a manager, customer or whatever may dictate the business).
As a software developer you should always have (at least) one of the prime factors under your control.
That way you might be able to devilver a product on time, in budget and at the required level of quality - but probably not with all the features on the wishlist.

'Design by commitee' is often a synonym for 'bad design'. So much is true. Design by team sounds mucht better (to me).

Happy developing - anyway

Stephan

[Peter Wone]

Permit me to clarify two points.

Mutability of specification does not, in and of itself, make it impossible to deliver a product of high quality. However, mutable specs are most often coupled with immutable budgets and timeframes, and this combination does significantly degrade the end result.

Nebulous specifications, in my experience, are frequently used as a blind by

• incompetent users who don't want their worthlessness exposed by detailed documented scrutiny
• lazy users who know they won't be held responsible for project failure
• actively resistant users who would like nothing better than to see the project fail
• professional avoiders of responsibility (this type abounds in the public sector)

When these are the causes of the vagueness in the specification, trying to get better definition is like pulling teeth.

There is another cause of project failure that I forgot to mention, to which you allude, and that is contradictory requirements. This is particularly devastating when combined with design by committee. Often the politics of the situation are such that competing empires are determined not to risk significant success on the other's part, and they make effective use of contradictory requirements to bog down the development process while they try to get their own house in order.

[David Rawling]

I'd like to add a clarification on the nebulous specification idea.

A few things to consider:

• Most users don't really know what they want - I mean really know what they want.
• Most of those that know what they want cannot articulate it.
• Most of those who could articulate it are not "powerful" or "important" enough to give input. By this I mean the people who will use the application - why should they have input?!
• The 1 remaining user who can articulate and who gives input is drowned out by the project managers, project directors, project sponsors, client management, consultants, directors, designers and architects, all of whom "must be better at it" because they have the right clothes.

Thus the good specifications are lost in a world of uncertainties, guesses and catch-alls.

Based on this list it becomes obvious that most user-provided specifications will be nebulous because you don't hear from the people who know what they want, you're not allowed to go FIND them to talk to, and management insists on pretty graphics and frippery to show the big bosses that things are really happening.

As for mutable specifications ... these come about because it seems modern managers cannot (perhaps "will not"?) make decisions. By this, I mean that most of the managers I know and have worked with are incapable of choosing one item from a list of two, even when the choices are clear cut, vis-a-vis:

• I wish to die a thousand deaths
• I wish to receive $1 billion tomorrow, retire and do what I want for the rest of my life.

(I should note that I have been privileged to work with 2 managers in my 7 years with my current employer who could, in fact, make decisions. The other 9 managers of varying types matched my description above).

What this leads to is the management droid in question specifying either:

• Both (Contradiction)
• Neither followed by a shouting match ("It's not in the spec" ... "Don't give me what's in the spec, give me what I want")
• One, then the other, then others (Mutability, unfocused change and lack of clarity for developers)

As Peter reminded us ... loose, mutable and nebulous specs usually arrive with tight, immutable and well-defined deadlines and budgets.

[drxenos]

When I was first looking at editors, before I settled on TextPad, I checked out Ultra-Edit. Nice editor, but I was turned-off by the overt religous tone of their website. I mean, to each his own, but what does God have to do with selling me an editor?

[Peter Wone]

If you'll forgive me for a deliberate misquote,

Ask not what God has to do with selling you an editor, but what an editor has to do with selling you God.

Religion is a pathological meme. It is [sic] catholic in its choice of vector.

[zridling]

Quick point on Ian Mead at UltraEdit - I agree the godboy stuff is a little over the top, but it's mostly a story (or ultimatum to his god) that if he didn't sell 20 registrations to his UltraEdit editor, he would stick with his full-time job over just developing UE full-time. Apparently his god took the dare and gave him 22 registrations and viola! UE was commercially born. Peter nails the relevance of such an anecdote regarding text editors.

I don't think anyone can say it's a bad editor, but it doesn't have the style and refinement that TextPad does.
___________________
David said, "Most users don't really know what they want ...and if so [they cannot articulate it]" or are not in a position to express their wishes; meaning, they're never allowed behind the closed/locked doors to the devs. And David, as you well know, what those same managers tend to do when they finally make a decision is deny it to their death, sidestepping as much responsibility as possible.

Just like UltraEdit, there seems to be a small number of people 'actually' working on TextPad. But isn't this forum and the direct line to support the vehicle to combat the point David is making?

I may express some wish or want or even detail a problem, but there are others here [in this forum] who can understand, outline, and express what I'd like to say far better than I could and in far more exact terms. But to me that's like having my spelling corrected - nothing shameful in that - just better to get it right, as folks like David, Peter, Stephan, and others have learned over the years how to refocus vague ideas into actuality.

[drxenos]

If you'll forgive me for a deliberate misquote,

Ask not what God has to do with selling you an editor, but what an editor has to do with selling you God.

Religion is a pathological meme. It is [sic] catholic in its choice of vector.

I don't understand.

[Wade]

It's in reference to:
http://www.ultraedit.com/products/story.html

[drxenos]

What do you guys think of this editor?

[skaemper]

What do you guys think of this editor?

Which one do you mean?

Both (TextPad and UltraEdit) have pros and cons. I personally have paid for both, but I curently use only TextPad.
It's clean, fast, does what I need, and works straight forward (almost always).
TextPad's not so strong points are

• Macros aren't editable
• Syntaxt highlighting doesn't allow for bold, italics and similar
• No Linux/Unix versions avaliable - [/i]native ones I mean
• No scripting (relates to the Macros mentioned above
• Regular Expression support is a bit weak. (compared to the RegExes of Ruby, Perl, Python, and other editors like emacs...9

As for UltraEdit it's mostly the UI: I think it, er, could be improved. I just don't like it (and therefore don't use it).
I consider this a hint to every software developer: If there's a major reason why users don't use your software or complain about the software: Go fix it. Now.

My €0.02

Stephan

[skaemper]

BTW, two other editors I know of are
Edit plus and TSE Pro (resp. Jr).
I used the semware editor to write my thesis back in the old days and was really happy with it. Today it may be a bit outdated due to the 'dosish' user interface.

Stephan

[Guest]

Which one do you mean?

Sorry, you must not have noticed my subject header. I should have stated it explicitly in the text. I was wondering what people that of MultiEdit.

[skaemper]

Sorry, you must not have noticed my subject header. I should have stated it explicitly in the text. I was wondering what people that of MultiEdit.

Oooops, sorry for not reading carefully.
MultiEdit, has probably a lot more features than the other editors mentioned here (up to now).
As far as I can tell, the main differences are

• It's not shareware
• It significantly more expensive (> $ 120)

And the last one makes a whole lot of a difference.At least it would make a difference to me (& my wallet ). But it doesn't because I think it's too expensive. (if I had to use a really extremely powerful editor I'd likely use, coughahem, emacs )

Stephan

[zridling]

Stephan's right: if you're a nonresponsive developer, then your product will continue to wither, as people who need certain features will go elsewhere to find THOSE specific ones. UltraEdit is a great editor, but with a lousy Word Count dialog and ugly icons. TextPad could use some more customizability features, but that's another post.

[RhinoBanga]

Plus TextPad could do with editing via FTP which is the only reason why UE is still on my machine.