Holy hell. An entire site, by hand? Who does that!
Wow. It’s like, he’s Neo, and this is the matrix. We all see this world and he sees 01010101001010 all over the place.
Crazy.
— Stevos 
As far as I know, yes. Certainly the weblog part of it is maintained by hand; he’s said that several times.
— Mark
“Holy hell. An entire site, by hand? Who does that!”
Is that actually so unbelievable? Once you get a good template going, its just a matter of plugging in content. I still think there is a fair amount of hand-coders…er…hand-mark-upers out there.
Out of curiosity, what do you people use anyways? I’ll sometimes use a WYSIWYG editor to throw together a basic template, but after that its hand-mark-uping all the way, and most of the time its hand-mark-uping from beginning to end for me. Now an RSS feed by hand…hmm…that’s another story altogether…
Zeldman does bring up an interesting point about how presentation can actually be a part of the “content” of a page.
— MikeyC
“Zeldman does bring up an interesting point about how presentation can actually be a part of the “content” of a page.”
I agree with Zeldman on this one. I have tried various RSS readers, but failed to see the point of reading sterile posts. Somehow they are more lively and carry part of the author’s personality when viewed within the site design.
— RobbyB
It’s totally unbelievable. MOST web sites today (that are updated fairly frequently) use some kind of automated content management. Even if it’s just a php script to put the content of a page in a box.
I didn’t mean hand coding as in BBEdit, I meant as in nothing server side doing work for you. Especially considering this is a weblog, and is therefore updated frequently!
Plus, forget the template thing, he pushed the idea that since every page is individual markup, you might have totally different schemes for each page. It’s insane!
But it is, on the other hand, worth complementing.
“I have tried various RSS readers, but failed to see the point of reading sterile posts.”
I’ve never really viewed RSS readers as an end-user type of tool. Just figured that RSS feeds were more suited towards use on small-screen devices or content aggregation something like this: http://www.readinged.com/ .
But isn’t RSS just a stop-gap measure on the way to the true semantic web anyway? If the semantic web ever does materialize on a wide-scale, then it would sort of make RSS redundant, wouldn’t it? But if Zeldman is right, then doesn’t that sort of throw a monkey-wrench into what the semantic web is envisioned as?
— MikeyC
“MOST web sites today (that are updated fairly frequently) use some kind of automated content management.”‘
I must be in the stone-age then as I maintain a site for a company (my day job) which is completely hand-mark-uped static pages (except for our product search which is obviously server-side and dynamic). And this is a fairly big company with about 20,000-50,000 unique visitors a month.
— MikeyC
“Zeldman does bring up an interesting point about how presentation can actually be a part of the “content” of a page.”
I’m been thinking about this and how I use RSS as a subscriber. I use NetNewsWire Lite to aggregate a number of feeds for me. Some of the feeds supply a summary, some only a headline. When I see an article I like, I follow the link to the article itself and view it in it’s natural environs–content and presentation.
Dive into Mark is the only site I subscribe to that includes a lot (all?) of the article in the feed, meaning I wouldn’t have to follow the link, but I find I do anyway.
I used to use OmniWeb’s “Check for changes…” and automatically check changes bookmarks feature as a way to find out when sites were updated. This had it’s problems: if I left OmniWeb up when I unplugged from the network (as I often do, carting my machine around from place to place), OmniWeb would mark all links as dead when it checked for updates. Also, some sites featuring ads would be marked as updated when the ad changed.
NetNewsWire Lite does a much better job of letting me know if I want to visit a site to read new posts, and saves me time because I don’t have to check pages just to see if they’ve been updated. By following the links provided by the feed, I get to read the article and see the presentation. Seems to me a great thing, for authors and readers alike.
Am I missing something? I may very well be, and await enlightenment.
— grzm
RSS is useful, tho not a replacement for the site itself. It provides a pull-y implementation for notification of web updates.
“But isn?t RSS just a stop-gap measure on the way to the true semantic web anyway?”
Godwin’s Law.
http://diveintomark.org/archives/2003/03/19/the_road_to_xhtml_20.html#c000656
RSS is useful -today-. If the SW comes to be, people will use it. But until then, we’ve got too many sites to read. ;)
— Jeremy Dunck
i think rss on the sites should be of the complete post. a line or two does not tell you anything.
— abhi
Writing the RSS feed by hand may be tedious, but just look at those summaries — they’re real *summaries*, not the result of
summary = posttext[:100] + ‘. More…’
Of course, you could include such summaries as a field of your post in your weblogging tool of choice. But how many people take the effort to condense their post, anyway? Maybe the mere fact of having to write it all by hand invites to a bit more dedication.
And after reading the summary for “There is no ping”, I’m afraid I’ll from now on have to check *both* the website and the RSS summaries not to miss anything ;)
Cheers
— Afonso
I think Zeldman.com is using a little bit of automation: Server-Side Includes (SSI), as noted by the *.shtml extentions on most of his pages.
I don’t think it detracts from his “hand-rolling” though.
MikeyC:
“I maintain a site for a company (my day job) which is completely hand-mark-uped static pages”
Learn a simple templating system, you’ll save at least 30% of worktime. Enjoy. (And tell nothing to your boss about it, of course).
— mur0
And now, the weather, in which Hell has frozen over.
— Jesper
“Holy hell. An entire site, by hand? Who does that!”
I do. In fact I have always done so. Like Zeldman, Tantek and others, I actually find it pleasurable (perhaps even theraputic) to code my web pages by hand. I messed around with Dreamweaver once and found it inflexible and cumbersome.
It alarms me that XHTML 2.0 is getting more and more complex because it is going to make it harder to code by hand.
I used to use server-side includes, but I found that for some reason their use slowed down the speed at which my pages were retrieved from a server, and caused Bobby to choke, so I decided to stop using them.
I hand code my RSS feed as well. I find this a little tedious (especially since I am completely new to XML).
As I have said before, I agree 100% that RSS is just a stopgap measure while people are still trying to grasp the Semantic Web. Keep in mind, however, that HTML is just a stopgap measure while people are still trying to grasp SGML.
RSS’s usefulness is immediately obvious to anyone who sees NetNewsWire (or whatever) for the first time. I’m less convinced about the current trend of embedding XHTML into RSS directly, or embedding RSS into other formats. If it solves a problem, great. And I’m still trying to figure out why I should care about RDF. (I haven’t given up hope yet, though. Shelley’s book will be out soon.)
— Mark
Hand-coding is one reason why Zeldman is such an avid standards supporter. Reworking every table, graphic, font tag, etc on an entire site with as many pages as Zeldman’s has is insane in comparison to editing a single stylesheet.
Hand-edited isn’t that rare, they even have their own clique – http://www.murky.org/handedited/
Personally, I redesign too often to hand-edit every page.
— Aquarion
The other issue, undiscussed here, is that some of us know HTML (*and CSS*, thank you, jfournier) but cannot program PHP- or whatever-based CMSes. My two least-updated, least-read Weblogs use a certain British Columbian/French Web nabob’s “back end,” but everything else is written by hand.
I don’t like it much, but at my current level of expertise, it’s either do it this way or fail to publish. And like I’m gonna do *that*.
#1: “Holy hell. An entire site, by hand? Who does that!”
I do. Other than my personal website (which is just a kind of journal for me), I work on a few intranet websites, all by hand.
Well… I do have some UltraEdit templates, but I have never found a content management system which seems easier to use than just typing it all in.
And as long as you do not use scripting garbage, and stick to XHTML and CSS, why not?
— Jor
I do my site by hand. It’s basically just a buildatemplatethencopyandpaste job.
— Menc
Zeldman writes his own RSS. It’s like the parting of the Red Sea, especially after reading Unsyndicate [1], which I don’t entirely agree with.
RSS has different usefulness for different people. Let the reader decide how they want to use it.
I subscribe to 44 RSS feeds and use Ranchero Software’s brilliant NetNewsWire (http://www.ranchero.com/netnewswire/) to check them all; my main reason for doing this is to save time.
I love the fact that I can launch a single application and have it download all the new content on the web sites that I enjoy reading; I do not have to trawl through each page manually, wasting time on web sites with no new content.
Of course, this means that I miss out on design etc. and I only get the raw contents of an update, but I tend to design my own web sites to be as simple as possible, making accessibility to the actual content the number one priority rather than concentrating on making pure gimmicky aesthetics. Therefore, I see no shame in not visiting a site to read it, instead prefering to pore through articles in the comfort of NetNewsWire.
— mudge
I tend to do a little of both using either bbedit or homesite. I hand code all of my pages using php, xhtml and css and then use php on the server to plug them all together. My weblog code on my site is hand written by me and I often update and change the various “parts” sidebar.html, topnav.php, mysql, bloginclude.php, etc. The only automated part is the blog posting bit and even then I do some markup to the post. Using php includes is a bit like using css. If I need to change something in the footer I can just edit footer.php if I need to change the layout just edit the css. I tried using one of those pre made blogger systems like b2 or mt. But after having to read through each part to find what I wanted to change I just decided to write my own.
Zeldman had a q&a one time that he answered the syndication question with something like “Why not just leave your food out on the front sidewalk so I can eat it with out coming inside?” I guess now he justs puts a snack out front leading us in.
I don’t hand code my site because I have my own CMS that i use for maintaining my site.
And some of you have mentioned that u didn’t find any thats good, have a look my journal, if you like it i’ll upload a zip.
(btw: its done in ASP)
— Abhi
The best thing about using CMS is that u need to make changes to just a basic layout. and thats surelot better that editing numerous individual pages.
I use FeedReader but the only problem is that the RSS contains only a short description Or the first few lines of the blog.
— Abhi
I code my site by hand. I’ve used CMSs before (Blogger, The Journaling Script) and, while they were fine, they’re not worth the trouble as I see it. I like the fine level of control over archiving (IMHO, most modern blogging tools have braindead archiving features). I like seeing the code. And I don’t see a loss in productivity. Installing MoveableType takes time. Getting all the templates to work together takes time. Typing your <p>s and <div>s by hand doesn’t really take much time at all.
— Eli
When [X]HTML gets too hard to edit, a simpler markup language (HTML 4? ;-) will rise to front-end it.
— Ken MacLeod
Now Zeldman just needs to stop forcing my browser to open offsite links in new windows and I’ll have an excuse to read the site regularly. Nothing like stepping on a landmine and getting a new window when you don’t expect it :P
— Tony
I’m not sold on having to dish out 10k every hour to every user running an aggregator. Even though my page is bloated and graphics-heavy, the 100k hit is still lighter than dishing out 240k/day per user, ignoring image caching.
http://diveintomark.org/archives/2002/10/21/push_by_any_other_name.html pretty much says it all.
— Dave S.
I used to hand-code my entire site (you can root around at archive.org if you care to see it). I would post a new article, oh, about once a month. I used BBEdit, I used include files, I was doing most of the smart things to minimize my workload.
Then I started using Blogger, and quickly switched to Movable Type. Now I post about once a day, and perhaps once a month I have a long post on the order of the monthly posts I made before.
Is it really *that* much easier with MT? No, but sometimes a small difference in inputs results in a big difference in outputs. A threshold is crossed. Not to mention the benefit that a CMS makes it easy to slice-n-dice your content in many different ways. And my website is much better, IMO.
So I have a mixture of admiration for Zeldman’s energy and meticulousness, and bemusement at what seems like an arbitrary insistence on doing things the “old-fashioned” way (can we say that about websites?) for no apparent reason.
Oh yeah, as to RSS: My website is not such a thing of beauty that I mind if people consume it via RSS. I prefer reading mosts blogs in RSS form (except for those few that are so pretty they need to be seen in their native form), and wish more people would publish fulltext feeds. I can read much, much more in NetNewsWire than I ever could in a browser.
An excerpt-only feed is still a good idea, though, since blog-reading in an aggregator isn’t the only use for a feed. I’m trying to put more energy into writing good excerpts.
I also hand code all of my work. Always have.
We are currently in the process of migrating the structure of several thousand objects I have managed by hand for years to a back end system which will automate all of it.
I am looking forward to it because it will free up my time to hand code a new project which I will no doubt build out and manage until the tools have been developed to automate it.
“I’m not sold on having to dish out 10k every hour to every user running an aggregator.”
Then what does this say about the likelyhood of the semantic web coming to pass? What insentive do companies/individuals have for moving towards a format that can be sliced and diced at will using robots without having to actualy visit the site and for their hardwork they pay for it with their bandwidth without opportunity to recoup it through banner ads/pop-ups etc?
Given the current attitude of fear and loathing towards deep-linking, how willing will companies be towards making their information infinitely more accessible through the semantic web?
Just questions, nothing more.
— MikeyC
Dave S,
If you are sending 100K every hour to anybody running an aggregator, something is going wrong (or you are updating more than once an hour).
HTTP handles things like this with caching. 304 Not Modified is your friend.
— Anonymous
excerpt or full text?
Why not a third option – a link to a super stripped down content chunk, possibly even XML encoded with all that extra namespaced overload. This way the RSS file is light and fast, and if someone wants to read more they can download it, but unlike the html web page it isn’t chock full of images/blogrolls/navigation/blogshares/picture-of-cat.
All we need now is the XML format of the content chunk (possibly even RSS?) and a namespaced RSS-mod to qualify the link.
LazyWeb, I invoke thee.
MikeyC,
”
What insentive do companies/individuals have for moving towards a format that can be sliced and diced at will using robots without having to actualy visit the site
”
I have a feeling that by the time SW comes along, Internet2 will also be rolling out. Before too long, we’ll look back on piddly little T1s and snicker like it was the 640k limit. But that’s not really the point.
”
and for their hardwork they pay for it with their bandwidth without opportunity to recoup it through banner ads/pop-ups etc?
”
Software is a service industry in factory’s clothing. Ads for eyeballs on the web are nowhere near the money makers they used to be. Frankly, SW offers companies with legitimate services unimaginable advertizing power– that is, ready accessibility to anyone looking for their product (or information related). That beats the heck out of a banner ad.
”
Given the current attitude of fear and loathing towards deep-linking, how willing will companies be towards making their information infinitely more accessible through the semantic web?
”
The web is architecturally an info interchange system. Companies that use it purely for marketing do so at their peril. I think the hubub about deep-linking will die down if/when the idea of retrievable resources is understood by people putting things out on the web.
There is no “site”, there is no “page”. There is only domain and resource.
I’ve seen a number of sites that practice security by obscurity– that is, they put out an unsecured URL, and then don’t link to it.
Honestly– my company has a product catalog online, with related applications. The marketing people didn’t want too many products on a single page, for fear that competitors might steal our product line info. However, it is a simple thing for a web UA to slurp all our pages programmatically, if someone were so inclined.
Hell, before the web, they could stroll into any of our dealers stores and walk off with a catalog, too.
That’s not to say that all stuff on the web should be freely accessible to everyone– HTTP does provide for authorization, and I imagine that as identity infrastructure becomes more mature, so too will the auth functionality.
Clearly, the uptake speed of the SW will be nothing like the commercialization of HTML, for several reasons: it’s not as visually sexy, dot-bomb memories, and social adjustment, to name a few.
It is quite right that the web is a fiercely competitive market, but I also feel that in time, a company desiring global channel reach will advertise its wares most effectively using the SW.
Some day. ;)
And, last bit– RDF/XML may not be the answer that catches on, but in the long view, that doesn’t matter. We’ve got a global communication infrastructure, and the promise of global KR is too huge to fail at forever.
— Jeremy Dunck
“The web is architecturally an info interchange system. Companies that use it purely for marketing do so at their peril. I think the hubub about deep-linking will die down if/when the idea of retrievable resources is understood by people putting things out on the web.”
I guess it depends on the reason why you have a web site to begin with. A company selling products online would probably not be threatened by the Semantic Web as easy, widespread aggragation and indexing can only help getting your products exposure and, thus, sales.
A site like TVGuide.com, might feel extremely threatened by the Semantic web. If I can simply pull my channel listings from their site without visiting, they lose the opportunity to bombard me with ads. So, in this way, I think that ‘difficult-to-index’ tag soup, would actually be preferable to the semantic web in their POV.
The Semantic Web/XHTML working groups seem to be envisioning a world of tomorrow that will somehow mirror the days before the mainstreaming of the internet when it was used largely for academic and scientific purposes. XHTML, mathML, SW, all seem about 10-12 years too late in my estimation. Not that I’m knocking them, as it will be great if they actually go anywhere, but I am very sceptical.
— MikeyC
“Thanks for finally making an RSS feed! I had stopped visiting your site because of its lack of one.”
This is a quote posted in http://www.zeldman.com/daily/0403a.shtml#soldout from an email zeldman received. This is not the first time i’ve seen (read) this attitude (certainly among bloggers anyway)in the last couple of months and i find it a little disturbing that a site may only be deemed worthy if it has an RSS feed.
As to what appears in the ‘description’ tag it obviously(?) depends on how you intend the feed to be used. An ‘i’ve updated’ feed requires no more than a taster of each post. any other use requires the full post i reckon. Also i’ve started to handroll my own feed and have found that i can make the description far more interesting than if it were auto-generated.
I wonder if we’ll reach a point where it’s normal for a ’site’ to simply consist of an RSS feed?
— pete
Pete,
It’s not a question of “worthiness”. Everyone only has a certain amount of time to spend surfing blogs. Visiting a hundred blogs to see what’s new takes time. Subscribing to them all, and being notified when they update does not take much time at all. You can keep up with more websites when they use RSS – it’s just efficiency.
— Anonymous
> Now Zeldman just needs to stop forcing my browser to open offsite links in new windows
Tony,
If you are using a Gecko based browser there is a pref to fix that:
// Windows open in new window instead of tabs (target=)
// http://bugzilla.mozilla.org/show_bug.cgi?id=105547
user_pref(“browser.block.target_new_window”, true);
— Phillip
*most* “web design” is *bad* design that interferes with the content. Zeldman notes some sites that, when stripped down to text in the rss feed, one notices “that they’re not well written” – great, now you can stop wasting time reading them. Assuming, of course, that you’re reading for information, and not entertainment…
Ping.
— Anonymous
Hand code *everything??* Why? I don’t have as much time as that to even *read* all of that blog.
I like includes, automation. Say, computers are good at automation, aren’t they?
Anyway, my hands hurt.
| Now Zeldman just needs to stop forcing my browser to open offsite links in new windows
I use Galeon (under Linux) and use ‘Open link in new tab’. Works great.
In fact I browse the web this way: go to a site with lots of links e.g. http://www.linux.org.uk/cgi-bin/portaloo
and then I do ‘Open link in new tab’ on every link I find interesting. Then I start reading all the new tabs and closing them as I go. The browsing is not linear anymore it’s more like a three.
Maybe Opera for Windows http://www.opera.no has a feature like this?
— Po