Sad Sigil news?

[Hitch]

Originally Posted by Katsunami

Columns in CSS....

I think I'm going to hang myself. If there's one thing that's hard to do in CSS, it's columns. Yeah, it can be done with tricks, but you often need(ed) different tricks for different browsers.

We no make columns. Customer want columns, customer can a) change mind, b) decide to do whatever it is in image, or c) decide that my harangue about single-columns is the right thing to do, or d) for smaller items, use a table. In ebooks, columns are just...a big NO. This is a constant discussion about indices, which have other issues, as well (how does someone with 3 entries in the index that reference, say, Winston Churchill, make sure that they get to and from the RIGHT index item?) Nope, nich, nein, nyet.

Being a coder, I wouldn't expect it to be otherwise, to be honest. "Could you please change <insert one little detail here>" can sometimes take a lot of work, no matter how modular and perfectly the application is designed.

No, of course not. Hell, I have to make wee alterations to my simple website all the time, and it's never just..one thing. It's ten things or whatever. Either we all put our heads together, and figure this out for Sigil, or I can pretty much guarantee that this will be The End. I don't see why we think that more people are going to come along and contribute to Sigil that aren't here already. Doesn't that make sense? That if they were interested in ePUBs, they would have already found us, here in our wee niche of the Netverse? Don't mean to be Dora Doom, but I think finding user_none was a miracle in the first place, and has worked out to be such. Sigil is tres fabulous now, and it would be nice if it could stay that way.

Again: just my $.02.

Hitch

[Katsunami]

I wonder how people created good EPUBs before Sigil... Just converting a Word document or exporting one through Writer mostly doesn't do it too well without extra cleaning.

[Hitch]

Originally Posted by Katsunami

I wonder how people created good EPUBs before Sigil... Just converting a Word document or exporting one through Writer mostly doesn't do it too well without extra cleaning.

Well, I can't speak for anyone else, but I simply did then, very much what we do now; I cleaned all the html in NoteTabPro. I use Sigil primarily as a finishing tool, as it saves making the NCX, the OPF, etc. And drag-and-drop for reordering sections, etc., is extremely handy and useful. I like the new reports function, and clips are nice, too (although, obviously, we have that in NTP). What Sigil does shouldn't be underestimated, though--it saves a ton of time in the NCX and OPF alone, if naught else.

n.b.: I also do a lot of fast pre-cleaning in Word, which then produces very clean code. Word produces ugly code when people use it poorly; once cleaned and with Styles used, you can take Word's output, remove the internal SS, and roll and go in HTML. If you export ad hoc-typed Word to HTML, then, yes, the clean-up is dreadful. But that's not Word's fault--that's the user's fault.

Hitch

[Toxaris]

Originally Posted by Katsunami

Being a coder, I wouldn't expect it to be otherwise, to be honest. "Could you please change <insert one little detail here>" can sometimes take a lot of work, no matter how modular and perfectly the application is designed.

We have had these discussions also on these forums. They even said, 'it will only be a small change and easily done'. When I argumented that it might be a big change, they basically said I was crazy, because they know it better. Some of these 'requests' were really a lot of work with very little gain.

[Toxaris]

I do not know C++, otherwise I might have been able to help out. I have the coding experience, but not the language and from what I understood the language can be cumbersome.
I am afraid that I have to second Hitch in this. The changes that additional coders will be found is slim in the next few months. User_none was indeed already a very big surprise and he made Sigil from a nice tool to a great tool.
If it was not for Sigil, I would still be coding ePUBs manually...

[user_none]

There are a few things I want to clear up.

1)

Originally Posted by st_albert

.. but since the source tree has been removed from google code ...

The source moved. The complete history was retained as part of the move. If you have a git checkout already from Google Code you can use the following to change the origin location you pull updates from:

 git remote set-url origin  <NEWURL>

We'll have to see how GitHub pans out but hopefully it works out better than Google Code.


2)
Monetary compensation isn't what I'm looking for. I have a full time job which I like more than Sigil. I don't want to make Sigil into a full time project and I don't want to deal with supporting a paid product (I get to do that already at work because we offer direct to developer support). The issue isn't money it's time constraints between Sigil and other things I'd like to do.

3)
Commercialization of Sigil. This is allowed by the GPLv3 license Sigil's code is released under. That said the GPL makes turning a project into a paid product very difficult. You can sell a package but the GPL requires you provide the source, and build files so anyone you sell the product to can build it themselves. It also says that you can't prevent someone from doing the same thing you're doing.

So selling binaries won't work. All that needs to happen is one person pays then they are allowed to re-distribute the binary as well as all source and build instructions to anyone. This is precisely what Red Hat deals with in regard to CentOS.

What Red Hat and other companies do is provide paid support. They also provide fee based development services. They also have bundling deals with hardware manufactures. That's how a commercial GPL project typically make money. Other open source licenses have different constraint and can be managed in a different manner in regard to commercialization.

The biggest headache really is support and I'd rather not get more involved in that. Especially as noted people are entitled enough when it comes to something they were given for free, once they've paid that entitlement sky rockets.

4)
The choice of the C++ language has been a hinderance to Sigil. It's a complex language and few people know it well enough that they're comfortable contributing. Pretty much every person whose contributed has said something like: "I know language X but C++ is confusing, is this patch correct?"

When I took over Sigil from Valloric he basically told me that if he knew what he knows now about how C++ would impact Sigil he would have used a different language for the project.

I'd think with Microsoft pushing C++ as their "system" language and embracing C++11 more people would be comfortable with it. But as it turns out C++ really is preventing contributors.

[Hitch]

Originally Posted by user_none

There are a few things I want to clear up.

1)


The source moved. The complete history was retained as part of the move. If you have a git checkout already from Google Code you can use the following to change the origin location you pull updates from:

 git remote set-url origin  <NEWURL>

We'll have to see how GitHub pans out but hopefully it works out better than Google Code.


2)
Monetary compensation isn't what I'm looking for. I have a full time job which I like more than Sigil. I don't want to make Sigil into a full time project and I don't want to deal with supporting a paid product (I get to do that already at work because we offer direct to developer support). The issue isn't money it's time constraints between Sigil and other things I'd like to do.

user_none: I knew that, but in my discussions here, I was really talking about finding other "contributors"/coders for Sigil's ongoing health. I hope you didn't think I meant you were "holding out" for ducats, that's not what I intended at all. :-) OTOH, a laborer is worthy of his pay.

3)
Commercialization of Sigil. This is allowed by the GPLv3 license Sigil's code is released under. That said the GPL makes turning a project into a paid product very difficult. You can sell a package but the GPL requires you provide the source, and build files so anyone you sell the product to can build it themselves. It also says that you can't prevent someone from doing the same thing you're doing.

So selling binaries won't work. All that needs to happen is one person pays then they are allowed to re-distribute the binary as well as all source and build instructions to anyone. This is precisely what Red Hat deals with in regard to CentOS.

What Red Hat and other companies do is provide paid support. They also provide fee based development services. They also have bundling deals with hardware manufactures. That's how a commercial GPL project typically make money. Other open source licenses have different constraint and can be managed in a different manner in regard to commercialization.

The biggest headache really is support and I'd rather not get more involved in that. Especially as noted people are entitled enough when it comes to something they were given for free, once they've paid that entitlement sky rockets.

Which is where my head blows up. This is the primary reason that if we fork Sigil, I wouldn't make it available. That's not our gig--neither software development nor any type of tech support. And, to be blunt, if I have to pay for something to maintain Sigil to a level that keeps up with the ePUB world, then, well...{shrug}.

4)
The choice of the C++ language has been a hinderance to Sigil. It's a complex language and few people know it well enough that they're comfortable contributing. Pretty much every person whose contributed has said something like: "I know language X but C++ is confusing, is this patch correct?"

When I took over Sigil from Valloric he basically told me that if he knew what he knows now about how C++ would impact Sigil he would have used a different language for the project.

I'd think with Microsoft pushing C++ as their "system" language and embracing C++11 more people would be comfortable with it. But as it turns out C++ really is preventing contributors.

Oddly enough, we have a C++ coder in-house, but she hasn't used it in a while. Nor do I think she'd be willing to dive into this. I suspect that to find someone may take some doing. If any of the other contributors to Sigil read this thread and are interested in possible work on a fork, please PM me. (Mods: I sincerely doubt that this violates any rules, but if that sentence does, please let me know.)

Hitch

[Katsunami]

Originally Posted by user_none

The biggest headache really is support and I'd rather not get more involved in that. Especially as noted people are entitled enough when it comes to something they were given for free, once they've paid that entitlement sky rockets.

True. However, I can understand that if someone pays for a product and there is a problem (and I don't mean "Teach me how to use this and that feature", I mean a real problem/bug), then it would be expected that it is fixed as soon as possible.

The choice of the C++ language has been a hinderance to Sigil. It's a complex language and few people know it well enough that they're comfortable contributing. Pretty much every person whose contributed has said something like: "I know language X but C++ is confusing, is this patch correct?"

When I took over Sigil from Valloric he basically told me that if he knew what he knows now about how C++ would impact Sigil he would have used a different language for the project.

I'd think with Microsoft pushing C++ as their "system" language and embracing C++11 more people would be comfortable with it. But as it turns out C++ really is preventing contributors.

I wonder.

C and C++ are not that difficult. I can learn a programming language within a few days, but then it's still not possible to write meaningful programs. The problem is not C++, or whatever language: the problem is (IMHO) the bazillion number of libraries.

I know C, C++, C#, PHP, and other languages too, but it's impossible to just jump into a project because most of the time, I don't know the libraries that are used well enough.

Doesn't Sigil use BOOST, QT, and some other libraries? I know BOOST (the few parts I've used of it), but I know nothing about QT, or any other library Sigil uses.

The language is not difficult, but if someone expects you to sit down and start writing code for some huge project: good luck. There are like 20.000 frameworks and libraries out there, and they're all different, and work in a different way. It costs time to get up to speed with the ones used, and after you know them enough to start coding, the knowledge becomes almost useless as soon as you need to work on a project that uses different frameworks and libraries.

I don't know if using a language other than C++ would have solved that problem.

[Hitch]

The language is not difficult, but if someone expects you to sit down and start writing code for some huge project: good luck. There are like 20.000 frameworks and libraries out there, and they're all different, and work in a different way. It costs time to get up to speed with the one used, and if you know it, the knowledge becomes almost useless as soon as you need to work on a project that uses different frameworks and libraries.

Yes. That's precisely the problem.

WRT Tech support expectations: I have fairly VAST experience with this, and 90+% of it is, "teach me how to use this feature" or "I can't be bothered to read the instructions." Just today I was castigated by a woman who had her book scanned and OCR'ed, even though I warned her, because I didn't teach her how to use Word beforehand, so that she could find and fix broken paragraphs. And I mean, CASTIGATED, as if this were part of my job.

Nothing could induce me to offer support for Sigil. Seriously. Look at all the entitled a**holes we get on here complaining, that we see now, when it's FREE????

Hitch

[crich70]

Originally Posted by Hitch

Yes. That's precisely the problem.

WRT Tech support expectations: I have fairly VAST experience with this, and 90+% of it is, "teach me how to use this feature" or "I can't be bothered to read the instructions." Just today I was castigated by a woman who had her book scanned and OCR'ed, even though I warned her, because I didn't teach her how to use Word beforehand, so that she could find and fix broken paragraphs. And I mean, CASTIGATED, as if this were part of my job.

Nothing could induce me to offer support for Sigil. Seriously. Look at all the entitled a**holes we get on here complaining, that we see now, when it's FREE????

Hitch

My dad always used to say, "some people would complain if they were hung with a new rope." Some people are never happy Hitch.

[Hitch]

Originally Posted by crich70

My dad always used to say, "some people would complain if they were hung with a new rope." Some people are never happy Hitch.

LOL, Crich:

I'm pretty sure I'd complain were I hung with a rope, new or used, Crich. I like that saying, I may appropriate it!

Hitch

[Katsunami]

Originally Posted by Hitch

Yes. That's precisely the problem.

That is exactly the problem, and many companies (even IT ones) don't seem to understand. When looking for jobs, it often went similar to this:

"So, you have C++ / PHP / C#... experience?"
- "Yes, but I've been mainly writing code for micro-controllers, in pure C for the last 2 years. I haven't touched PHP or any of the frameworks for at least that long."
"OK, we'd like you to do a little test, in PHP."
*point*
"There's a computer, and an assignment describing a function we would like you to implement in our <insert huge on-line administrative package here>, within 2 hours."

Dude. That's not going to work, and YOU FRACKING KNOW IT.

No wonder all those IT companies can't find any engineers. Many of them expect you to sit down and write code, as if it is the same as boxing and wrapping toothbrushes in a factory.

It's the same with those who'd like "An all-round IT-expert": you know, the people that know PHP, Python and Perl for their CMS, C++ for a few old desktop applications, C# for some newer applications, Bash scripting in Linux to manage our Linux servers, and oh... you must be able to also manage Windows servers and Oracle databases on the side.

Believe me, I've seen those kinds of job offers. Often.

You won't find someone who can do all that this at once at a professional level.

Oh, and with regard to tech support: you're quite right. I've enough experience with it. I want *this* feature to be on any new computer, to be remotely controllable ->

[st_albert]

Well, I am, or used to be, fluent in C as well as several other languages, but I was never able to grok C++ for more than a couple of hours at a time, on a good day. It wasn't the language itself, I'm ashamed to say, but the Object Oriented paradigm itself that I didn't find intuitive, what with its constructors, destructors, overburdened operators... I don't know what all.

IIRC there's a very famous C++ comment that goes something like this:

// when I wrote this code, only God and I knew what I was doing.
//  now, only God knows.

Yeah, you knew where that was heading right from the first line, yes?

Kind of like bad Fortran or Basic spaghetti code. If you don't comment the hell out of it as you go along, three months later you'll never understand it. Even if you wrote it yourself. You needn't ask how I know this. ;-)

So, yeah, the chance of finding one or more C++ coders who can dive into a million and a half lines of code and do anything useful in terms of major revisions (epub3, anyone?) anytime soon would be miraculous. And it's already happened once.

Albert

[crich70]

Originally Posted by Hitch

LOL, Crich:

I'm pretty sure I'd complain were I hung with a rope, new or used, Crich. I like that saying, I may appropriate it!

Hitch

Feel free. Dad alas has been gone 17 yrs now, but his wisdom lives on.

[Greg Anos]

Originally Posted by Katsunami

That is exactly the problem, and many companies (even IT ones) don't seem to understand. When looking for jobs, it often went similar to this:

"So, you have C++ / PHP / C#... experience?"
- "Yes, but I've been mainly writing code for micro-controllers, in pure C for the last 2 years. I haven't touched PHP or any of the frameworks for at least that long."
"OK, we'd like you to do a little test, in PHP."
*point*
"There's a computer, and an assignment describing a function we would like you to implement in our <insert huge on-line administrative package here>, within 2 hours."

Dude. That's not going to work, and YOU FRACKING KNOW IT.

No wonder all those IT companies can't find any engineers. Many of them expect you to sit down and write code, as if it is the same as boxing and wrapping toothbrushes in a factory.

It's the same with those who'd like "An all-round IT-expert": you know, the people that know PHP, Python and Perl for their CMS, C++ for a few old desktop applications, C# for some newer applications, Bash scripting in Linux to manage our Linux servers, and oh... you must be able to also manage Windows servers and Oracle databases on the side.

Believe me, I've seen those kinds of job offers. Often.

You won't find someone who can do all that this at once at a professional level.

Oh, and with regard to tech support: you're quite right. I've enough experience with it. I want *this* feature to be on any new computer, to be remotely controllable ->

You left out being about to read COBOL....