h2g2 The Hitchhiker's Guide to the Galaxy: Earth Edition

Echoing a comment made under the < / B R > conversation, I'd also like to express my condolences at the loss of off-site images. It's made my own personal space terribly dull.

I've nothing against the joint venture with the BBC and I appreciate the reasons behind removing off-site material, but is there any way we could see some kind of new image service where we can request or generate our own on-site stuff?

From what I read elsewhere, I think the long-term plan is a kind of 'BBC hosts the image' service, with all the usual legal shenanigans.

Shame it's gone -- I only ever linked to images that I own the rights to (and in fact created) -- ie: logos for my online Web stuff...

Let's face it, I'm no literary genius- My homepage might as well be empty now that I'm not even allowed to display my own drawings on it.

My tiny sketches were the only thing that ever attracted any feedback from other visitors, so I won't be coming back to the new H2G2 until the situation is resolved- And preferably *not* by signing over the image copyrights to the BBC.

Awww...
Give it a second chance, eh?

Well, maybe I'm just overreacting. Still, it did come as a bit of a surprise- I don't really understand why we can't just have a disclaimer whereby people state that images used are their own.

Surely any 'offending' images displayed after this could have been removed via the normal complaints system anyway?

I'm sure there's a whole conversation thread somewhere that's already covered this, probably Peta or Abi could point us toward it if they ever see this.

But basically, while you've just made a very good point, I know that that must have already been discussed during the past five weeks, and I'm certain that there's a very good explanation why that can't be done.
Server speed or something? Alignmant with the moon on Tuesday evenings? I dunno...

I'm sure you're right, but of course as a mere mortal I haven't had access to any of these conversations until today, so I'll carry on ranting here in case anybody picks up on it.

I appreciate that the BBC wouldn't want to offer their own servers for H2G2 images due to speed/bandwidth costs, but this doesn't explain why we can't use externally linked images like before. In my case, and I assume most others, my images are stored on my own webspace which I have paid for(as part of my ISP package). Where's the harm in that?

If it's a legal issue, then why are moderators more likely to be able to spot copyrighted text than copyrighted images?

No idea, 2b honest. Thou appearst more savvy than I in the technical department.

i spent hours finding new images & updating my page from my Word program & i got it looking (in my head) brilliant with new images & everything & its so amazingly dull now....all writing & no fun

Hi pyramid!

Yep - there has been such a conversation (or six) already. Basically, the key is...

Copyrighted text can generally be discovered by simply doing a websearch for the text of the entry. If any hits turn up large chunks of similar text, clarification will be required. This doesn't cover books, etc - but one hopes that few people will be bored enough to write large chunks of stuff word for word from a book, so it's not a major issue.

Images currently have no easy mechanism for finding out the copyright holder. Proposals such as digital signatures, etc, may change this in the future, but currently this is a bit of a showstopper.

Worse, an offsite image can change without h2g2 being aware of it. Such a change could put a piece of child porn, say, directly on h2g2's pages, if the moderators checked it while it was OK, and it was subsequently made non-OK. That would put h2g2 in breach of all kind of laws, and in severe danger of being taken to court. Mark Moxon could end up in jail for several years.

In addition, there's the issue of 'bandwidth theft' - it's not a good idea to use lots of off-site images - because the hosters of the images tend to complain and/or block their use.

All these things have been a problem 'to be resolved' for some time. But the switch to the BBC, and the subsequent predicted influx of new researchers, has made it rather more urgent.

h2g2 are giving bandwidth for hosting of the 'h2g2 life' series by Wowbagger, for example. Try talking to Shazz and asking if you can make your sketches into a similar series for the post. I don't know what the copyright situation would be, though. I think you'd get to keep it, but I'm not totally sure.

http://www.bbc.co.uk/h2g2/guide/thepost

Hope this Helps

I know there are problems with Corrupt images ect and i am perfectly happy not put them on as they always slowed down the loading time of h2g2...its just i had really nice pics (like Smilies, band photos, even a shakespeare gif i had to look everywhere for, for an entry i was writing!) and its how long it took me too get them & rewrite my page around them & all the HTML coding & stuff which i have now had to scrap and still can't get my page to come up without errors!

It just seems a shame seeing peoples hard work creating fascinating, interesting, colourful pages go down the plug hole! If you get what i mean!!

Em

Yeah, it is a shame. Such is life, sadly.

Best option is a page elsewhere, and a link to that page. /For pics of the Basingstoke gig, click here/

It isn't the same, though...

Thanks for all the replies everyone... However, I'm still not convinced.

> Copyrighted text can generally be discovered by simply doing
> a websearch for the text of the entry. If any hits turn up large
> chunks of similar text, clarification will be required. This
> doesn't cover books, etc - but one hopes that few people will
> be bored enough to write large chunks of stuff word for word
> from a book, so it's not a major issue.

That *almost* sounds like a challenge.

>Images currently have no easy mechanism for finding out
> the copyright holder. Proposals such as digital signatures,
> etc, may change this in the future, but currently this is a
> bit of a showstopper.

Let's not forget that a successful search based on a section of text is no guarantee that it's copyrighted material or that the person who posts to H2G2 doesn't *own* the copyright. Will the moderators *really* be able to check text in this way? Which search engine would they use?

> Worse, an offsite image can change without h2g2 being aware of
> it. Such a change could put a piece of child porn, say, directly
> on h2g2's pages, if the moderators checked it while it was OK,
> and it was subsequently made non-OK. That would put h2g2 in
> breach of all kind of laws, and in severe danger of being taken
> to court. Mark Moxon could end up in jail for several years.

Again, I'm not sure that this is all that difficult to solve. Why not checksum the images at the time that they're submitted and hide them until they've been approved. Software could then be used to periodically verify image checksums and re-hide those that have changed. (Wanders off to patent the idea)

> In addition, there's the issue of 'bandwidth theft' - it's not
> a good idea to use lots of off-site images - because the hosters
> of the images tend to complain and/or block their use.

Yes, if people are using Geocities or similar, this is true. However, people can usually use their own ISP webspace for whatever they like, and we're unlikely to be talking about 1024x768 24 bit images...

> All these things have been a problem 'to be resolved' for some time.
> But the switch to the BBC, and the subsequent predicted influx
> of new researchers, has made it rather more urgent.

Fair enough...

>h2g2 are giving bandwidth for hosting of the 'h2g2 life' series
> by Wowbagger, for example. Try talking to Shazz and asking if
> you can make your sketches into a similar series for the post.
> I don't know what the copyright situation would be, though. I
> think you'd get to keep it, but I'm not totally sure.

Hmmm, interesting idea. I might just do that! (Wanders off to look at the H2G2 life series)

>Why not checksum the images at the time...

But the external images aren't on h2g2 (& would never even pass through their servers normally - even when the page was displayed the image travels from it's server to you, not through h2g2).

To do the CRC calculations they'd have to load the images onto h2g2, then calc & store the CRC values. Then regularly reload the images, recalc the CRCs & compare to the stored value to check for changed images. Also identical CRC/checksum values are possible (though unlikely) for non identical images.

;^)#

Where is the difference between including a picture and including a sentence? If I type the letter 'a' it's an order to the browser to show this letter. If I type a -tag it's an order to the browser so show this picture. Both can offend (maybe with a certain context).

h2g2 has full control because it can remove both immediately. In order to find material that breaks the rules you have to look carefully anyway. You can't just use a search engine. I drew quite complex star maps for a constellation project -- now the articles look castrated and discussion with other contributors and h2g2 about including them into the Guide becomes more difficult.

Maybe every researcher should be able to reserve a picture disk space on h2g2. For me e.g. 1MB would be perfectly sufficient.

There is only one difference between the letter & an image url - once the letter is stored on h2g2's servers you can't change it. With an image url you can change the displayed image without changing the image url simply by substituting a new image file at the same address.

Today's starchart could be tomorrows Buffy vidcap with no change in the entry data at h2g2; ie undetectable by h2g2.

As the Beeb insist that h2g2 check all external URLs regularly the work load involved in maintaining some sort of OK'd image listing, that made the image identifiable, which was checked against the images currently displayed would be horrendous.

;^)#

Even the current system doesn't guarantee real security. To me this image rule is pretty paranoid.

I couldn't argue with that - but we should remember that the rule has come from the BBC rather than h2g2 staff.

;^)#