pouët.net

I know almost all the effort goes in to making the actual Hugi releases (getting articles, doing interviews, tweaking the layout and engine, etc.), and I appreciate that.

However, the HTML dumps look like nobody even tried. I'm sorry--not really--if this is insulting, but they're awful! They might be more readable without any styles at all. Pride and effort was taken in the releases themselves, why none there?

I don't think it would be a monumental amount of effort to make a simple styled page (template) that is used for all web releases, going forwards, to make them nice to look at and read. If this going-forward-template had nice markup, it would be easier to style it if perhaps each one should look slightly different. It would be nice if the existing 36 dumps with their really bad markup could be mended, but I rather think that might take a bit.

At any rate, I did a live mock-up in an hour or so (while my ISP decided the community didn't need internet for a while and this morning after it returned) of the Hugi 33 Table of Contents and the Making of "nano" article. I think it is much better, and I'm not anywhere near the best web developer. I tried to use colors and font choices from the original web dump, so it not completely different. (Though I did move one item in the table of contents so I didn't end up with a dangling single column item before a double column one.)

Any other thoughts? I can't be the only web guy who thinks "yuck".

Thank you for publishing this draft! I like it very much and would appreciate it greatly if you adapted all articles from all issues to fit this style. The HTML dump I created is really only a dump, for issues #18 - #37 it was made by saving the articles inside the Panorama engine to HTML (F2), and the articles from older issues were converted manually. I'm well aware of the weaknesses of what I put online and for this reason I greatly appreciate your support.

It takes a bit to convert one page, and those issues are huge :P I think it would take forever if I did it. I don't exactly have any magic to make convert from the broken HTML that was output, but I suppose it might not be that hard to make one that gets most of the big stuff done. (font tags with Century Gothic -> h3, adding closing paragraph tags ...)

I've uploaded two more pages, your Editorial for that issue, and the immediately proceeding article for the mean time. All the links on the contents page are now broke because I took out the base tag, so as pages fill in it should just work.

Maybe once one issue is done others can get an idea and help out with the remaining.

Well, web browsers are pretty good at parsing shit markup and making sense of it. You could make a simple crawler with a client part that extract the bits you want and saves the cleaned page/snippet somewhere.

They are! Due to all the legacy crap out there they have to handle. However, that legacy stuff isn't nice to style ;) I could maybe get part way there with what I've done for the new pages without a markup transformation, but I doubt I'd be able to get all the way there.

The one benefit, though, is that once it's been reparsed (or scraped) into something nicer formatted, styling it how you want is a breeze.

(also, another page up)

Yeah, it's the HTML markup that's the problem with the Hugi content - you nailed it.

I've finished conversion of the "Welcome!" and "Headlines" articles. This is not a fast process. I think I'm worn out for today, and this is only a small part of the mag.

I realized that a tool might not be very helpful, even if written, because there's a lot of things I decide to do based on the actual content regardless of the original markup (such as turning a flat comma-separated list of people into a ul).

At any rate, should be good to check out for now at least, until I get around to finishing the rest of it. :P (Write shorter articles! Just kidding :) Maybe I should cherry pick remaining short pages for conversion ...

I don't quite yet know what to do with the left-aligned images at this point. They're ones that pertain to a section in an article, and it seems odd if they're too wide, pushing the text too far to the right (unnatural to read). So for the moment I've made them as wide as those naturally are in the "Nano" article, though that means for articles where sections have text longer than the image it might look a little better if their corresponding images were allowed to be a little bigger (just over twice as large).

Images that don't directly pertain to a section but rather to the article as a whole, and aren't a large photo (which I let take up the full width), I right align and they don't (yet) have a width restriction as there's enough room on the left to read stuff naturally.

Edit, nevermind. It's just that one "Farbrausch Demomaking.." paragraph that makes me think that way (about the left-align images being too small now). It should probably be broken up in to two paragraphs ;)

Hi,

Left-aligned images usually have a width of 486 pixels. That is the width of one column inside the magazine. Therefore I have not explicitely centered them, since the engine displays them in the same way.

If a left-aligned image is smaller than that, it is really meant to be displayed to the left of the current paragraph.

Thanks for your effort! There is a little typo in the editorial: You wrote "Adko" instead of "Adok". Other than that, it seems to be fine!

Ah, well, on the web at least it seems to me to a be a bit odd to read if the text is pushed too far over to the right, so I've made left-aligned images smaller (as I noted above) but hopefully not too small. A script could probably sit at the bottom of the page and offer a lightbox for those larger than the CSS width they've been allotted ...

So far the only article that didn't really fit the left/right flow I'd set up was the Timberland article.


Oops! I generally copy the text from view source, I must have typed that one out. I've fixed it.

I've got most of the main articles done ... My next task though is to start work with a "tag soup" library (there's more than a few) to get a reasonable DOM be able to be able to set up some rules about transforming the patterns the engine seems to output. Then someone could go do another over on it and clean it up. Might save a little time, depending on how good the transforming rules are.

I'm still making some calls for certain things myself that it's easy for me to see, but it might be easier/quicker to add those changes after the transformer is done.

"Starstruck" by The Black Lotus (Hugi #33) is missing. Blank page.

you are wasting your precious time on Hugi.

panic, not to sound rude, but when he's editing Hugi html by hand, I guess his time is not that precious

uff, hugi/adok is shit jaja, blabla etc.

abstracting from adok's person etc, i still think there were at least some useful coding articles and interviews in there, which i welcome to have available online.

Less of it is by hand than when I started. There's an ugly script that converts the existing articles to something less wart-y. A chunk of articles (especially those without content images) are good to go and just need to be uploaded. Others it's just a few quick fixes in file (or additions to the script to handle the missed case) and rearrangement of images.

panic, hey man, chill out. Whether or not you like this diskmag has no bearing on my actions.

Truck asked me to take a look at PAIN, too, but I said "I'll have to see". If the data can be obtained in a nicer starting format, it might not be too hard to write a script that outputs things which need very little tweaking if at all.

For those counting, I've finished the first link in the Reviews section. All before that is done.

Hugi 32, and 34+ I believe have a similar layout to 33, so I don't have to rewrite the puller part of the script that gets all the files so if I just stick fix egregious problems and upload the results of the quick conversions a lot could be done faster. The same goes for 33 itself, and I may do that too depending. Making it through all the articles / article types will allow me to identify all the warts I'm trying to fix so the script can accommodate them.

I've added the missing review.

Thanks! Already converted and up.

for seeing something in the scene that needs improving, and actually doing something about it

Thanks gasman!

Also, last thing for this evening (or night), I had corrected files previously converted so they have a proper footer now and pages with images get a script to offer a simplistic lightbox. Images displayed in a space smaller than the source are clickable (cursor indicates) to activate this. I didn't spend a lot of time styling the lightbox, but it should do. Images on the left side are going to be ones that are able to be lightboxed. Those on the right or center aren't width limited.

spike: then i guess you would look the same thing to Adolf Hitler if you abstract his wicked personality? :o

hitler never published coding articles, im afraid.

I dont get it...what for do we need online-versions of DISKmags ?
If i want to read a diskmag, i just download/start the .exe and i have the whole thing as it was intended...correctly aligned gfx inside the articles i read, accompanied by chosen music that fits reading about the scene.
Even if its a diskmag on a platform != pc its like less than 10 clicks to start up an emulator, set it up and start the disk. ;)
This way i am guaranteed to get what i was looking for...it wasnt a book, it wasnt some html-page with css-style, it was the whole thing, the whole experience of reading a Diskmag!
Articles may be fine, but it only gets special with fitting gfx and music!
This is the Demoscene ;)

Right: go make a demo!

i am making more demos than most of us!
i did like 18 4ks in 4 years! :p