Why I'm not too concerned about the current #AIart

 

I don't consider these overhyped bitmaps to be magic or even that intelligent. When we see the results of these text to image services, we are looking at a fairly straightforward machine learning neural network application, albeit on a massively large training data set scale.

The data sets are so large and were scrapped together so quickly, without curation or even human review that they contain a lot of straight rubbish and often the darkest side of human behaviour. Whilst this is unpleasant it will be difficult to fix without starting again but there is a taint of not quite right. Importantly, a lot of the material used in the training was copyrighted but that was ignored as it was scrapped and fed into the neural nets. This is legally questionable, but the big money potential may spend this small problem away.

I think it is the lack of ethics in not attributing or even asking artists if they wish to contribute, that could be the undoing of the magic. If the majority of those consuming the images on the net don`t care or have already scrolled over the issue in their cosy little social media bubble, nothing will happen. However, if those looking also see the discussion of unhappy artists and appreciate the widespread collection and indiscriminant use of existing art renderings (style and colour, not cut and paste copies). Perhaps they might investigate and get a whiff of what is so on the nose. Let’s hope so.

A dozen regrets, and reasons I’m not excited about google photo

That 15^th March deadline is approaching fast for the cutting off picasa in favour of google photos. Although I know I must keep using google photo for my blog, but it is clearly wanting in many aspects compared with Picasa web albums, connected to Picasa of the desktop

1. Sharing privately is now very difficult if your family member do not have google accounts. It is possible to send links to selected photos and albums, and whilst you can share publicly to a number of social networks with one click, you need to get a link and then cut and paste to be able to email the link. The URL is not a direct link to the photo anyway but some google address that google needs to translate. All this is tedious and not obvious.
2. Albums in google photos should be called events or photo sessions as they are rigidly saved as the chronology (of the first uploaded photo). Something you are trying to constantly append to (eg a 365 daily photo or year long photos from one location) disappears into the past If you don’t provide an album organization the photos are grouped by date anyway. One small screen and even standard opens because of the album tile size, anything older than a couple of weeks disappears “below the fold” (ie you have to go scrolling to find it).
3. You can no longer choose create common licensing. In fact it would appear you can not copyright images either (at least there is nothing to that effect displayed). The terms of service
4. The People organization is not as comprehensive as the people (face recognition) in picasa
5. Classification of Places and Things are even less successful. What’s more disappointing here is there is no way to redress the misclassifications. (eg. my wife does not appreciate being on the tile labelled things)
6. Why oh Why! was SnapSeed editing removed from google photos? The current edit capabilities are so very limited. Its just sad really.
7. Why is something as fundamental as resize (eg smaller to email) missing?
8. Where is the exposure histogram?
9. What about all the other EXIF & metadata?
10. What has happened to geotagging (location)? Is it merely being used occasionally in the location organization. Why not have an option to display it in photo details?
11. I haven’t been successful in getting google photo autobackup to be selective, either from my phone camera or computer, it just behaves like a vacuum cleaner (and miss-behaves a lot as well, so I have removed it). The android google photos app kept updating itself and then crashing requiring yet another download and re-setup I just gave up and deleted it (and enjoyed having a lot of space back on my phone). If my older android version was not compatible wouldn’t it have been better not to have kept trying to update. No matter its gone also.
12. I’m frequently off the grid (ie no internet access a lot of the time) and I normally load and preview on my laptop computer first (I know I will continue using picasa to do this) THUS the [DEPRECATED] picasa connection makes the cloud based google photos irrelevant to me.

The list could go on but my firm opinion now is that google photos is way below my expectations. Ok Google photos can keep storing images for my blog but is not the place I intend to use to share and store my photos on-line anymore. I definitely intend to keep using picasa on my computers locally but the search is on for an on-line service that makes sharing and organizing easy, that will at a pinch do a reasonable job of simple editing, not strip out metadata and clearly display licensing, eg creative commons &/or copyright. Face recognition and auto organization might be nice but that’s not the deal breakers, lack of control is.

More Copyrights & Wrongs and Thinking About Editions

Create Your Own Provenance

My disappointment with Google (and most other social web services) at their ignoring creative commons still stands but I have been researching how Australian Computer Copyright law might be applied for Local Photographers. It is pretty simple if you are working for yourself (even free lance and you intend to sell your work) the images always belong to you, you don’t have to register your image, or display a copyright symbol on it or sign away the rights to some other organization or person. The copyrights always remain with you. You can licence the right to publish the image and this should normally be for a specific publication or purpose (ie use to advertise an exhibition) at any time. If you are employed to take photographs , the copyright belongs to your employer, Most of the guff about copyright you may read on the web really relates to the elaborate legalistic constructs to favour publishers (in the USA). Now the tricky bit is proving the image is yours (and this is where establishing provenance and the POSSE approach becomes important). Further when publishing on the net with “orphan works” laws being considered in some countries it doesn’t hurt to include a copyright or creative commons notice clearly visible within the post or site.

With that semi-rant out of the way, another photographic misconception that sit uncomfortable with me is that having a media that is easily copied makes the works less valuable. Such artificial marketing actions like limited print editions and destroying negatives, are the realm of gallery owner and art dealers and don’t benefit the creative person behind the image. There is only so much you can do and thus your output will always be limited. I recently came across the marvelous artist statement/manifesto of Brooks Jenson, which cover my sentiments exactly and his analogy to the printing of books and numbering of editions strikes a powerful cord.

I’m not sure how or why I have missed Brook’s wonderful LensWork for so long, it has a wealth of content covering much of the wider aspects of Photography and the Creative Process, and it is definitely worth bookmarking.

Bad Flickr …not bad panda

For some reason I believed that flickr had stopped stripping the EXIF data out of images when they are shared. This might not sound like a big issue but it is for those that rely on the image having an embedded copyright or creative commons entry in the IPTC and EXIF metadata. This in theory means that anyone that “finds” an image on the net should be able to trace the image back to its rightful owner. The trouble begins when various countries are enacting so called “instagram acts”, which allow “orphaned” photos to be freely used without and copyright or creative commons restrictions. Stripping the meta data away when and image is displayed effectively makes the image an orphan, if the user doesn’t want to take any further effort to find its owner.

I was experimenting with some of the sharing and social network integration in Irista, and happened to notice when irista bought images back from flickr the metadata was stripped. This surprise me because valid meta data is a key check that irista does on upload.

Flickr is only one of the popular services doing it. The only service I have found not doing it is google+

Looks like it is back to always including a visible watermark on flickr, or more probably not uploading important images there anymore.

Start your own Provenance, sharing into the future

Social media is often prompted as a way an artist can reach a larger audience. This is really only partly true. Just putting works on line doesn’t mean others will look at them.  For a start it will probably just be a tiny thumbnail to most until they click on it, if they click on it. Further in a time where almost everyone has access to the net, via smartphone, tablet or computer and their own web presence, via a social media service. Everything quickly gets swamped by a plethora of other distractions. Many of those distractions will be striking similar because the web is full of copy cats and people who “wanna bee” celebrities.  They either are ignorant of, or just wish to remain ignorant of, intellectual property rights and ownership, they are just chasing likes, views and +1s. If your work is original, you may not want it devalued by lots of “clones” or just simply having it “stolen” and reposted without attribution. Well maybe you don’t actually need to post it, just about it.

Establishing provenance (the chain of ownership) for art works is an important historical undertaking. It helps detect and eliminated forgeries (unless the forger can also “plant” an elaborate chain of fake histories) and it is fundamental in establishing the value of the work. It is not possible for you to write the provenance but things you do, write about, sketch and/or photograph will form the foundation of this future value. Posting about your current work, including sketches and photos of you working on it and/or others in front of the work (say at an exhibition) are probably better ways to use social media to promote your art than just posting a thumbnail of it.

The Short Black was painted in 2013, as a donation to the Art Auction at Alphington Primary School

Copyrights or Copywrongs

Whilst the majority of the social web is just ignoring the issue of personal IP (intellectual Property) creative commons have updated their licences. So what is Creative Commons about? Like its older (pre-net cousin Copyright the “All Rights Reserved” identity) to seeks to defined just how your intellectual property can be accessed and used. Un-like the copyright message it allows you to set the conditions under which your works (photographs, blogs, songs) can be share, reshared or reused. All in relatively easy to understandable English (or a lot of languages).

I have long championed using creative commons licencing when posting photos (and infact any content on the web).  Unfortunately it has not really helped me much in the two problems I have encountered, that’s more to do with the organizations involved, which are google & yahoo (via flickr). However I do see that creative commons is a very sound method. It is not time to throw the baby out with the bath water. So here for those that missed some earlier post is a short history of some bad behaviour on the net and lacl of action by those that should know better.

1. I received a take down notice via yahoo but supposedly from the DCMA to remove a photo mosaic of my own eye (shown above) claiming I had modified the original (owner was not identified) and reposted it. Well the photo mosaic was definitely my own (compare it to my buddy icon) and it was posted on flickr in my wallpaper set, back in 2007. It had also had over a few  thousand views and I did expect others to download it (for personal use on a screen saver) but even reposting it via flickr was fine as well, because the normal sharing process does embed a link back to your photo. Using Tineye I was able to find that there where a lot of repost of the image, several modified, infact one location was used as the background image for a notorious hacker cooperative. I was able to email them and have the image removed without any DCMA threat. Happy ending, seems there is honour in the Hacker community. However in trying to reply to Yahoo I got nowhere. If my emails where received they where ignored. Not such a happy ending, infact no ending.
2. For a while whole sections of my blog where being scrapped and reused in the faux camera review web sites (I won't give names as I don't want then to get any traffic from me). I tried tracking them down but it was a complex multi-national goose chase. So I embedded some "hidden links back to my stories". That was ok for a while but they soon wised up and deleted all my carefully crafted links. The actually acknowledge me by approximately the right name but with a link to what I presume where their sites. All these site were running ad-sense, so I tried to contact google, because they where presumable paying these guys for the traffic my content was attracting to these sites and that would definitely violate my non-commercial condition in my Creative Commons licence. Well despite approximately 9 months of trying a never got an answer, again I gave up getting no answer.
3. I’m very dissapointed that google+ photos no longer display the creative commons licence conditions for photos, although it is still possible to add such licence to an image in the underlying picasa web albums.

So why am I still recommending Creative Commons? I am actually convinced that over legalistic behaviour (particularly associated with publishers and record companies) has created the paranoia and misinformation about copyright, which has contributed to the general public just ignoring it and assuming anything on the net can be copied. I would prefer a world where I can share my creativity, knowledge and perhaps innovation, providing I am acknowledge as the original source and on the extra condition that my work is not sold by others for a profit. I like that others choose to do this too, we can share a better world. Luckily there is a way to notify the wider net community of these conditions using a creative commons licence (see below and on the right hand panel of this blog)

Wandering in the Light by imageo is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.

What is that watermark about?

I have started to physically (aka visibly) add a watermark to my photos onloine. Mainly those shared on Flickr, but now any large resolution photos I post. I started doing this when I read that many services where stripping out the metadata, effectively making any downloaded (or screen copied images) into “orphan works”, which under some legislation in different countries (not Australia I believe) means that the images can be freely reused. Most of the things I post on this blog I do allow to be reused under a creative commons licence.  So if you see this watermark on an image you want to use, please contact me, I will probably grant you permission. I know a lot of people don’t like visible watermarks and if that is what puts them off so be it. The real reason I'm posting the watermark here is to ensure that google image indexes this simple design. After going to the trouble of designing a icon style shape that incorporated a stylized camera, the copyright symbol and the letters api (for art and photography @ imageo) I realised this would mean nothing to the majority of people who of course don’t know me or follow this blog. At least by having the watermark graphic indexed and locatable for a reverse image search means that anyone who tries to “steal” or repost these photos without permission are not taking reasonable steps to find the rightful owner. Such miss-appropriation can expect at least a DCMA notice or appropriate legal action of the country concerned.

Now you see it, now you don’t

its a while since I’ve mentioned flickr, in fact i really haven’t been posting to it much. Not because of the new format (I’m not really a fan but its much the same as other offerings) its all about “stripping”. They are still stripping the EXIF & IPTC metadata (including details or copyright or creative commons) and the photo ownership details) when the images are downloaded. This unfortunately creates what are known as orphan works (photos that in theory you can not establish the photos ownership). The problem being that in some countries it would appear re-publishing such photos, without the copyright holders permission is acceptable providing the re-publisher acted in good faith in searching for the photos owner, even though that was unsuccessful! See the legal loops holes opening up here!!!

 

Flickr isn’t the only social media service to do this but they are the one I would prefer to use.

This actually posses a big problem for creative people who want to share their work, even using creative commons. For example I don’t mind other using my images providing they acknowledge me, and that they don’t use my images for their own profit. A lot of people have contacted me and all is sweat, Unfortunately using google image search or TinEye I know some of my images have been used all over the place without my permission. So I have reluctantly taken the approach of adding a visible watermark onto my flickr posts. In reality this is not going to stop unauthorised reposting, or help me track down those just stealing the photos. To do that I should just stop posting to flickr (or most other social web sites). I hope it will just make those thinking of using eth photo think twice before downloading it. Flick actually has a good blogging link feature that includes a link back to the original flickr image (which does have the EXIF data intact).

If you want to use one of my images for a blog, or brochure for your organization, or any other reasonable use and you don’t like watermarks just send me an emails

Blimey, …they do that? ..aka(time for a manifesto).

I had heard the story about a new law, known as the instagram act, in the UK making “orphaned” photos free from normal copyright restrictions on republishing. So how does a photo become orphaned? Most jpegs, the most common format for posting photos on the web, have important detail about which camera, its setting and who took.owns the photos embedded in the photo itself, These are the EXIF and IPTC metadata and because they are embedded within the photo they travle with it when it is copied. Whilst some older programs may not handle them correctly, programs like picasa and lightroom and many others help you expand the details stored here. However then when I started to investigate some of the recent social photo sites I came face to face with the fact that most of the social sites where stripping out the metadata. In fact of the sites I have used only google+ is not altering the metadata. I must admit I was pretty shocked that Flickr was one of the worst offenders.
Right at the moment, posting your photos on social web sites, may not protect your rights much despite anything they might suggest in their terms of service. Add to this this “overlooking” of creative commons licencing on google+ and the web has become a less secure place to “share” your creations. Fortunately there is some good guidance being prepared.
IPTC Photo Metadata Working Group have prepared five guiding principles as our "Embedded Metadata Manifesto":

1. Metadata is essential to describe, identify and track digital media and should be applied to all media items which are exchanged as files or by other means such as data streams.
2. Media file formats should provide the means to embed metadata in ways that can be read and handled by different software systems.
3. Metadata fields, their semantics (including labels on the user interface) and values, should not be changed across metadata formats.
4. Copyright management information metadata must never be removed from the files.
5. Other metadata should only be removed from files by agreement with their copyright holders.

Avoiding the Black Hats & Spammers

I have become aware that at least one of my blog posts has again been “scrapped” and reposted as an exact copy of my original post, including picture and layout, on a spam site (I will not give you the link because you should not go there, and specially never every click on links you find on these dodgy looking “content farms”). At worse they will contain links to other spam and potentially malware. At best these lazy spamdexers are just trying to attract innocent links, thus page views and a higher google ranking, to their site using lots of plagiarized (aka stolen) content. It is a dishonest practice, and sorry I can do nothing about it.
What disappoints me most about all this black hat SEO trickery and blog scrapping is the fact that Google in many cases is paying the very spammers and thieves for the stolen content! The content farms have mostly been greedy & also monetized their site using Google’s own adsense. So try reporting this Google??? I have found there is a massive blankness, no response from anyone. You may remember my previous post on a lesson of theft versus sharing. Well that site, which looks a lot like a legitimate camera review site, still exists, it just keeps changing its name slightly, its ISP, its country of origin and its owner name, so that DMCA notices to cease and desist also go unanswered. At least that low life now acknowledges my posts and links back to my site (he may think technically he doesn’t breach my Creative Common Licence, but what he does is still not legal because he is expecting Google to still pay him for my content and my CC license is for NON commercial).  I can not find him, but Google obviously still can because he still displays an adsense badge. Am I being too cynical about Google’s apparent hypocrisy and will to fix this? Rant over for now.

As I indicated above there is little I can do, but there might be something you can do if you use chrome as your browser, Google has released an add-in that gives you a personal blocklist of such sites as you come across them. It will also transmit your blocking patterns back to Google, in theory they will then get a good crowd sourced blacklist of such websites and thus downgrade these undesirables in their famous search algorithms. Let us hope Google will follow through on this.

It is written

I have been photographing the interesting patterns of life, shell trails, as below, crab balls and diggings, footprints, etc on the sand for some time. I don’t don’t how to read their hidden message, when the next tide is, or is the swell dropping, or is a big flood likely, or is it just dinner time. Or is there even a message there at all?

I am finally succumbing to watermarks on my images uploaded to the net. I know a lot of folk hate them but there is just so much open theft and disregard of the rights of those posting on the net. I once championed the idea of the creative commons system that was adopted on flickr. Now I see the same images reposted so often I’m saddened. The copyright doesn’t mean I wont give permission to use and repost my work, but please write to me first. There are lots of utilities to add watermarks to your photos, many free and it is a common feature in photo mangers. I’m using the new feature to add a watermark included at the bottom of the file export in Picasa 3.0. It is simple and unobtrusive.

Play me a song - about the future

Play me a song, originally uploaded by wolfcat_aus.

I went along to pubcamp in melbourne yesterday, looking at where Web 2.0 might go next and was clearly the oldest bloke in the audience (fortunately richard walsh was involved in the debate so I was not the oldest overall)
The real timetracking of twitter hasshtags was impressive, and I guess the general consensus was content is king, but it is "hard to make a buck" (Stephen Mayne) and the advertising networks have the money in their pockets already.

So where does that leave the humble content creator? Do you sell your soul to google (yahoo! etc) or can you keep your digital rights and expect a a tiny bit of value to flow back to the content creator. Or is the general perception going to remain, if its on the net, its free to copy (and reuse).

If you want to know more ben barren's blog is much more eloquent than mine.

Special applause to Jed White for organizing the event.