[GOAL] Re: Fwd: The Open Access Interviews: Paul Royster

Sun Sep 21 08:54:05 BST 2014

As a layperson I would certainly be interested to know what margin of error
levels we can assume the “Web of Science and/or in Scopus” approach has. I
am conscious, for instance, that some of the reports by UK universities into
RCUK compliance mention using Web of Science, but they all appear keen to
stress that they have serious concerns about data accuracy. 

A list of RCUK compliance reports, by the way, can be found here:
http://goo.gl/Yi3twT

There is also a very informative blog post on the topic of monitoring open
access mandates/policies by Cameron Neylon here: http://goo.gl/Y02S87

Richard Poynder

From: goal-bounces at eprints.org [mailto:goal-bounces at eprints.org] On Behalf
Of Jean-Claude Guédon
Sent: 20 September 2014 23:27
To: Global Open Access List (Successor of AmSci)
Subject: [GOAL] Re: Fwd: The Open Access Interviews: Paul Royster

Extremely good answer, Bernard!

It is also very good to clarify the fact that the 90% figure is calculated
against the baseline of a combined WoS_Scopus search. However, and this was
part of my difficulties with Stevan's argument, I suspect that in SSH, in a
French-speaking university, many publish in French-language journals that do
not appear in either list. This means that, for Liège, the baseline works
from one year to the next, but if you want to compare Liège's mandate and
its effectiveness (which, once again, I agree, is - from common sense - the
best) with another kind of mandate in an English-speaking university, the
baselines will not be comparable.

If, furthermore, you imagine two universities that not only differ
linguistically, but also differ in the relative weight of disciplines in
research output - say one heavily slanted STM and the other heavily slanted
SSH, this too will affect the baseline simply by virtue of the fact that SSH
publications are not as well covered by WoS and Scopus as are STM
publications.

In conclusion, the baseline is OK for comparisons of a mandate's
effectiveness longitudinally, of for comparison purposes of two successive,
but different, mandates, assuming the institution remains pretty much the
same over time in terms of mix of research emphases; it is far more
questionable across institutions, especially when different languages are
involved (but not only).

Incidentally, what proportion of papers deposited in ORBI do not appear in
either WoS or Scopus? That too would be interesting to know as it might help
Stevan refine his baseline and thus make it more convincing.

Finally, given that all universities require, an annual assessment of
performance, including a bibliography of publications in the completed year,
would it be difficult to compare the repository's holding against the
publications of the researchers as declared by them? Knowing researchers,
every last little scrap of paper will be minutely listed in the yearly
assessment forms... 

-- 

Jean-Claude Guédon
Professeur titulaire
Littérature comparée
Université de Montréal

Le samedi 20 septembre 2014 à 19:10 +0200, Bernard Rentier - IMAP a écrit :

Dear Richard, 

Here are the answers: 

1. ORBi, the Liège University Repository, will soon (I believe) reach 90%
compliance. It is our target for 2014 and I hope we make it. 

This figure comes from the calculation of the percentage of ULg papers that
can be found in Web of Science and/or in Scopus that are deposited in ORBi
as well (see method in  http://eprints.soton.ac.uk/340294/) 

It concerns one year at a time and it is not cumulative. Last May, the
compliance level for the publications of 2013 was already 73% and our figure
for 2012 is in the 80% range. 

2. Only a small proportion of ULg papers are in CC-BY. 

This is simply because, in order to publish in the journal of their choice
(I haven’t tried to do anything against that!), our authors, in the great
centuries-old tradition, give away their rights to the publisher. We have no
control on that. 

Later on, there is no way for them to CC-BY the same text (in fact, we are
preparing ORBi 2.0, that will offer a CC-BY choice). 

For now, we are aiming at free access and we are not yet fighting hard for
re-use rights. We shall move progressively in this direction of course,
while the publishing mores evolve

In other words, I agree that we have free access, not a full fledge open
access yet. It is not a failure, it is our objective to gain confidence
first. 

Unfortunately, even if we have established in-house rules for evaluation,
external evaluations are still based on traditional indicators such as the
highly and rightfully criticized but widely used Impact Factor and the like.
In these conditions, today we cannot sacrifice our researchers — singularly
the young ones — in the overall competition for jobs and funds, on the altar
of « pure » Open Access. 

Best wishes 

Bernard Rentier 

Rector, University of Liège, Belgium 

Le 19 sept. 2014 à 21:52, Richard Poynder <ricky at richardpoynder.co.uk
<mailto:ricky at richardpoynder.co.uk> > a écrit : 

Dear Bernard, 

I have two questions if I may: 

1.       You say that Liège is getting close to 90% compliance. Can you
explain how you know that, and how you calculate compliance levels? I ask
this because the consistent theme coming through from UK universities with
regard to compliance to the RCUK OA mandate is that they simply do not know
how many research outputs their faculty produce each year. If that is right,
what systems does Liège have in place to enable it to produce a
comprehensive list of research outputs that UK universities apparently do
not have? 

2.       Does Liège track the licences attached to the deposits in its
repository? If so, can you provide some stats, especially the number of
items that are available CC-BY (which we are now told is required before a
deposit can be characterised as being open access? 

Thank you. 

Richard Poynder 

From: goal-bounces at eprints.org <mailto:goal-bounces at eprints.org>
[mailto:goal-bounces at eprints.org] On Behalf Of brentier at ulg.ac.be
<mailto:brentier at ulg.ac.be> 
Sent: 19 September 2014 18:46
To: Global Open Access List (Successor of AmSci)
Subject: [GOAL] Re: Fwd: The Open Access Interviews: Paul Royster 

"Liège does not mandate anything, so far as I know; it only looks into the
local repository (Orbi) to see what is in it, and it does so to assess
performance or respond to requests for promotions or grant submissions."
(JC. Guédon) 

Oh no, Jean-Claude, Liège mandates everything. 

It is a real mandate and it took me a while to get almost every ULg
researcher to realise that it is to his/her benefit.  

Linking the deposits to personal in-house assessment was the trick to get
the mandate enforced in the first place. As well as a few positive
incentives and a lot of time consuming persuasion (but it was well worth
it). 

Last Wednesday, the Liège University Board has put an ultimate touch of
wisdom on its mandate by adding "immediately upon acceptance, even in
restricted access" in the official procedure. Actually, a nice but to some
extent useless addition because, with time (the mandate was imposed in
2007), ULg authors have become so convinced of the increase in readership
and citations that two thirds of them make their deposits between the date
of acceptance and the date of publication.  

All this explains why we are getting close to 90% compliance, an outstanding
result, I believe.  

Le 18 sept. 2014 à 23:40, Jean-Claude Guédon
<jean.claude.guedon at umontreal.ca <mailto:jean.claude.guedon at umontreal.ca> >
a écrit :

A reasonably quick response as I do not want to go into discursive tsunami
mode...

1. Stevan admits that his evaluation of compliance is an approximation, easy
to get, but not easy to correct. This approximation varies greatly from one
institution to another, one circumstance to another. For example, he admits
that language plays a role; he should further admit that the greater or
smaller proportion of SSH researchers in the research communities of various
institutions will also play a role. in short, comparing two institutions by
simply using WoS approximations appears rash and unacceptable to me, rather
than simply quick and dirty (which I would accept as a first approximation).

The impact factor folly was mentioned because, by basing his approximation
on the WoS, Stevan reinforces the centrality of a partial and questionable
tool that is, at best, a research tool, not a management tool, and which
stands behind all the research assessment procedures presently used in
universities, laboratories, etc.

2. Stevan and I have long differed about OA's central target. He limits
himself to journal articles, as a first step; I do not. I do not because, in
the humanities and social sciences, limiting oneself to journal articles
would be limiting oneself to the less essential part of the archive we work
with, unlike natural scientists. 

Imagine a universe where a research metric would have been initially
designed around SSH disciplines and then extended as is to STM. In such a
parallel universe, books would be the currency of choice, and articles would
look like secondary, minor, productions, best left for later assessments.
Then, one prominent OA advocate named Stenan Harvard might argue that the
only way to proceed forward is to focus only on books, that this is OA's
sole objective, and that articles and the rest will be treated later...
Imagine the reaction of science researchers... 

3. Liège does not mandate anything, so far as I know; it only looks into the
local repository (Orbi) to see what is in it, and it does so to assess
performance or respond to requests for promotions or grant submissions. If
books and book chapters are more difficult to treat than articles, then
place them in a dark archive with a button. This was the clever solution
invented by Stevan and I agree with it.

4. To obtain mandates, you need either faculty to vote a mandate on itself
(but few universities have done so), or you need administrators to impose a
mandate, but that is often viewed negatively by many of our colleagues.
Meanwhile, they are strongly incited to publish in "prestigious journals"
where prestige is "measured" by impact factors. From an average researcher's
perspective, one article in Nature, fully locked behind pay-walls, is what
is really valuable. Adding open access may be the cherry on the sundae, but
it is not the sundae. The result? OA, as of now, is not perceived to be
directly significant for successfully managing a career. 

On the other hand, the OA citation advantage has been fully recognized and
accepted by publishers. That is in part why they are finally embracing OA:
with high processing charges and the increased citation potential of OA,
they can increase revenues even more and satisfy their stakeholders. This is
especially true if funders, universities, libraries, etc., are willing to
pay for the APC's. This is the trap the UK fell into.

5. SSH authors are less interested in depositing articles than STM
researchers because, for SSH researchers, articles have far less importance
than books (see above), and, arguably, book chapters.

6. I am not citing rationales for the status quo, and Stevan knows this
well. This must be the first time that I have ever been associated with the
status quo... Could it be that criticizing Stevan on one point could be seen
by him as fighting for the status? But that would be true only if Stevan
were right beyond the slightest doubt. Hmmmmmmmmmm!

I personally think he is right on some points and not so right on others. 

Also, I am simply trying to think about reasons why OA has been so hard to
achieve so far, and, in doing so, I have come to two conclusions: too narrow
an objective and too rigid an approach can both be counter-productive.

This said, trying to have a method to compare deposit rates in various
institutional and mandate circumstances would be very useful. I support
Stevan's general objective in this regard; I simply object to the validity
of the method he suggests. Alas, I have little to suggest beyond my
critique. 

I also suggest that  a better understanding of the sociology of research
(not the sociology of knowledge) is crucial to move forward.

Finally, I expect that if I saw Stevan self-archive his abundant scientific
production, I would be awed by the lightning speed of his keystrokes. But
are they everybody's keystrokes?

Jean-Claude Guédon

-- 

Jean-Claude Guédon
Professeur titulaire
Littérature comparée
Université de Montréal

Le jeudi 18 septembre 2014 à 12:28 -0400, Stevan Harnad a écrit : 

On Wed, Sep 17, 2014 at 10:53 AM, Jean-Claude Guédon
<jean.claude.guedon at umontreal.ca <mailto:jean.claude.guedon at umontreal.ca> >
wrote: 

Most interesting dialogue.

I will focus on two points:

1. Using the Web of Science collection as a reference: this generates all
kinds of problems, particularly for disciplines that are not dominated and
skewed by the impact factor folly. This is true, for example, of most of the
social sciences and the humanities, especially when these publications are
not in English.

The purpose of using WoS (or SCOPUS, or any other standardized index) as a
baseline for assessing OA repository success is to be able to estimate (and
compare) what percentage of an institution's total annual refereed journal
article output has been self-archived.  

Raw total or annual deposit counts tell us neither (1) whether the deposits
are refereed journal articles nor (2) when the articles were published, nor
(most important of all) (3) what proportion of total annual refereed journal
article output is deposited. 

Institutions do not know even know their total annual refereed journal
article output. (One of the (many) reasons for mandating self-archiving is
in order to get that information.) 

The WoS (or SCOPUS, or other) standardized database provides the denominator
against which the deposits of those articles provide the numerator.  

Once that ratio is known (for WoS articles, for example), it provides an
estimate of the proportion of total institutional article output deposited.

Anyone can then "correct" the ratio for their institution and discipline, if
they wish, by simply taking a (large enough) sample of total institutional
journal article output for a recent year and seeing what percentage of it is
in WoS! (This would obviously have to be done discipline by discipline; and
indeed the institutional totals should also be broken down and analyzed by
discipline.) 

So if  D/W, the WoS-deposit/total-WoS ratio = R, and w/s, the
WoS-indexed-portion/total-output-sample = c, then c can be used to upgrade W
to the estimate of total institutional article output, and the WoS deposit
ratio R can be compared to the deposit ratio for the non-WoS sample (which
must not, of course, be derived from the repository, but some other way!) to
get a non-WoS ratio of Rc.  

My own prediction is that R and Rc will be quite similar, but if not, c can
also be used to correct R to better reflect both WoS and non-WoS output and
their relative sizes. 

But R is still by far the easiest and fastest way to get an estimate of
institutional deposit percentages. 

(As far as I can see, none of this has much to do with impact factor folly.
For non-English-language institutions, however, the non-WoS correction may
be more substantial.) 

Stevan has also and long argued about limiting oneself to journal articles.
I have my own difficulties with this limitation because book chapters and
monographs are so important in the disciplines that I tend to work in. Also,
I regularly write in French as well as English, while reading articles in a
variety of languages. Most of the articles that are not in English are not
in the Web of Science. A better way to proceed would be to check if the
journals not in the WoS, and corresponding to deposited articles, are
peer-reviewed. The same could be done with book chapters. Incidentally, if I
limited myself to WoS publications for annual performance review, I would
look rather bad. I suspect I am not the only one in such a situation, while
leading a fairly honourable career in academe.

Authors are welcome to deposit as much as they like: articles, chapters,
books, data, software. 

But OA's primary target (and also its primary obstacle) is journal articles.
Ditto for OA mandates. 

All disciplines, including the social sciences and humanities, in all
languages, write journal articles. This discussion is about the means of
measuring the success of an OA self-archiving mandate. It applies to all
journal articles (and refereed conference articles) in all disciplines. 

There are problems with mandating book deposit, or even book chapter
deposit, so that is being left for later. 

Nothing is being said about performance review except that the way to submit
journal articles should be stipulated to be repository deposit. 

2. The issue of rules and regulations. It is absolutely true that a
procedure such as the one adopted at the Université de Liège and which
Stevan aptly summarizes as (with a couple of minor modifications):
"henceforth the way to submit refereedjournal article publications for
annual performance review is to deposit them in the [appropriate] IR ". 

Liège does not mandate the deposit of books. 

However, obtaining this change of behaviour from an administration is no
small task. At the local, institutional, level, it corresponds to a
politically charged effort that requires having a number of committed OA
advocates working hard to push the idea. Stevan should know this from his
own experience in Montreal; he should also know that, presently, the Open
Access issue is not on the radar of most researchers. In scientific
disciplines, they tend to be mesmerized by impact factors without making the
link between this obsession and the OA advantage, partly because enough
controversies have surrounded this issue to maintain a general feeling of
uncertainty and doubt. In the social sciences and humanities where the
citation rates are far less "meaningful" - I put quotation marks here to
underscore the uncertainty surrounding the meaning of citation numbers:
visibility, prestige, quality? - the benefits of self-archiving one's
articles in open access are less obvious to researchers, especially if they
do not adopt a global perspective on the importance of the "grand
conversation" needed to produce knowledge in an optimal manner, but rather
intend to manage and protect their career.

I am not sure what is the point of the above observations. I agree it has
been difficult to get authors to deposit. That's why the OA movement has
turned to mandates, and now to ways of optimizing mandates so as to
facilitate and maximize success (i.e. deposit rates). And here we are just
talking about how to measure and compare those deposit rates between
institutions, and between mandates. 

Nothing about impact factors, performance evaluation criteria, metrics,
discipline criteria or language differences. Just ways to induce journal
article authors to deposit them in their institutional repositories.  

Saying all this is not saying that we should not remain committed to OA, far
from it; is is simply saying that the chances of success in reaching OA will
not be significantly improved by simply referring to "huge" benefits at the
cost of only a few extra keystrokes. This is rhetoric. The last time I
deposited an article of mine, given the procedure used in the depository I
was using, it took me close to half an hour to enter all the details
required by that depository - a depository organized by librarians, mainly
for information science specialists. All these details were legitimate and
potentially useful.  However, while I was absolutely sure I was doing the
right thing, I could well understand why a colleague less sanguine about OA
than I am might push this task to the back burner. In fact, I did so myself
for several months. Shame on me, probably, but this is the reality of the
quotidian.

I invite Jean-Claude to time me depositing an article in my institutional
repository (and I am not a fast typist)! It takes about two minutes. 

In conclusion, i suspect that if Stevan focuses on such a narrowly-defined
target - journal articles in the STM disciplines - this is because he
gambles on the fact that making these disciplines fully OA would force the
other disciplines in the humanities and social sciences to follow suit
sooner or later. Perhaps, it is so, but perhaps it is not. Meanwhile,
arguing in this fashion tends to alienate practitioners of the humanities
and the social sciences, so that the alleged advantages of narrowly focusing
on a well-defined target are perhaps more than negatively compensated by the
neglect of SSH disciplines. yet, the latter constitute about half, if not
more, of the researchers in the world. 

The target is journal articles in all disciplines. Not clear why SSH journal
article authors would be any more or less compliant with self-archiving
mandates than any other discipline. It has nothing to do with books, yet. 

Yes, once journal articles are being self-archived universally, many other
things will follow. 

I suggest that it may be more constructive to practice deposit keystrokes to
provide OA than to cite a-priori rationales for the status quo, Jean-Claude.
I bet you'll be up to speed after depositing just a few articles! 

Stevan Harnad  

Le mercredi 17 septembre 2014 à 07:07 -0400, Stevan Harnad a écrit :

Begin forwarded message:

From: Stevan Harnad <harnad at ecs.soton.ac.uk <mailto:harnad at ecs.soton.ac.uk>
>

Subject: Re: The Open Access Interviews: Paul Royster 
Date: September 16, 2014 at 5:28:48 PM GMT-4

To: JISC-REPOSITORIES at JISCMAIL.AC.UK
<mailto:JISC-REPOSITORIES at JISCMAIL.AC.UK> 

On Sep 16, 2014, at 2:46 PM, Paul Royster <proyster2 at UNL.EDU
<mailto:proyster2 at UNL.EDU> > wrote: 

At the risk of stirring up more sediment and further muddying the waters of
scholarly communications, 
but in response to direct questions posed in this venue earlier this month,
I shall venture the following 

Answers for Dr. Harnad

(1) What percentage of Nebraska-Lincoln output of peer-revewed journal
articles (only) per year is 
deposited in the N-L Repository? About 3 months ago I furnished your
graduate student (at least he 
said he was your student) with 5 years of deposit data so he could compare
it to Web of Science 
publication dates and arrive at some data-based figure for this. I cautioned
him that I felt Web of 
Science to be a narrow and commercially skewed comparison sample, but I sent
the data anyway. 
So I expect you will have an answer to this query before I do. If the news
is good, I hope you will 
share it with this list; if not, then let your conscience be your guide. As
for benchmarking, I don’t believe 
it is a competition, and every step in the direction of free scholarship is
a positive one. I hope when 
they hand out the medals we at least get a ribbon for participation. 

Thanks for reminding me! It was my post-doc, Yassine Gargouri, and I just
called him to ask about 
the UNL results. He said he has the UNL data and will have the results of
the analysis in 2-3 weeks! 

So the jury is still out. But many thanks for sending the data. Apparently
Sue was not aware that UNL 
had provided those data (and I too had forgotten!). 

(2) Why doesn’t N-L adopt a self-archiving mandate? 
I do not even attempt to explain the conduct of the black box that is my
university’s administration; 
so in short, I cannot say why or why not. I can only say why I have not
campaigned for adoption of 
such a mandate.  My reasons have been purely personal and idiosyncratic, and
I do not hold them 
up as a model for anyone else or as representing the thinking or attitude of
this university. Bluntly, 
I have not sought to create a mandate because I feel there are enough
regulations and requirements 
in effect here already. Instituting more rules brings further problems of
enforcement or compliance, 
and it creates new categories of deviance. There are already too many rules:
we have to park in 
designated areas; we have to drink Pepsi rather than Coke products; we have
to wear red on game 
days; we can’t enter the building through the freight dock; etc. etc. etc. I
simply do not believe in 
creating more rules and requirements, even if they are for our own good. The
Faculty Senate 
voted to “endorse and recommend” our repository; I have not desired more
than that. But I am 
concerned mainly with 1600 faculty on two campuses in one medium-sized
university town—not 
with a universal solution to the worldwide scholarly communications crisis.
I see discussions lately 
about “putting teeth” into mandated deposit rules, and I wonder—who is
intended to be bitten? 
Apparently, the already-beleaguered faculty. 

I agree that we are over-regulated! But I think that doing a few extra
keystrokes when a refereed 
final draft is accepted for publication is really very little, and the
potential benefits are huge. Also, 
there is some evidence as to how authors comply with a self-archiving
mandate — if it’s the right 
self-archiving mandate, i.e., If the mandate simply indicates that
henceforth the way to submit refereed 
journal article publications for annual performance review is to deposit
them in UNL’s IR (rather than 
however they are being submitted currently) then UNL faculty will comply as
naturally as they did 
when it was mandaed that submissions should be online rather than in hard
copy. It’s just a technological upgrade. 

(3) Why do you lump together author-pays with author-self-archives?
I was not aware that I did this, so perhaps you are responding to Sue’s
catalog of various proposed 
solutions—“author-pays OA, mandated self-archiving of manuscripts, CHORUS,
SHARE, and others”—as 
all being “ineffectual or unsustainable initiatives to varying degrees.” I
feel we are strong believers and 
even advocates for author self-archiving (so-called), and disdainful
non-advocates for author-pays models. 
But I think we have become aware of the divergence of interests between the
global theoretics of the 
open access “movements” on the one hand and the “boots-on-the-ground”
practicalities of managing 
a local repository, even one with global reach, on the other. Crusades for
and controversies about 
“open access” have come to seem far removed from what we actually do, and
now seem more of a 
distraction than a help or guide. 

I can understand that, from the library’s perspective: The library can’t
mandate self-archiving,  can’t fund 
author-pays, and can’t do anything about authors’ rights. But maybe, if you
look at the evidence that 
mandates work, and become convinced, then the library could encourage the
administration
 And 
of course if self-archiving is mandated at UNL, then the library can help
with mediated self-archiving, 
at least initially, as I pointed out to Sue (though it’s hardly necessary,
for a few keystrokes — certainly 
a much smaller task than UNL’s current mediated deposit: tracking down the
PDF. checking the rights. etc.). 

We have been (and continue to be) constant supporters of “green” open
access; and we have appreciated 
Dr. Harnad’s reliably indefatigable defenses of that cause against
innumerable critics, nay-sayers, and 
“holier-than-thou” evangelists of competing approaches. I sympathize with
his weariness, I applaud his tirelessness, 
and I do not wish to tax his patience further. I hope no part of this
response will be interpreted as attempting 
to dispute, contradict, or diminish any of his points. I regret if these
answers are unsatisfactory or incomplete, 
but that is all I can manage at this time. 

Much appreciated, Paul!  

Hope to have the UNL data for you soon, with a comparison with other IRs,
mandated and unmandated. 

Best wishes, 

Stevan 

Paul Royster
Coordinator of Scholarly Communications
University of Nebraska–Lincoln
proyster at unl.edu <mailto:proyster at unl.edu> 
http://digitalcommons.unl.edu <http://digitalcommons.unl.edu/> 

_______________________________________________
GOAL mailing list
GOAL at eprints.org <mailto:GOAL at eprints.org> 
http://mailman.ecs.soton.ac.uk/mailman/listinfo/goal

_______________________________________________
GOAL mailing list
GOAL at eprints.org <mailto:GOAL at eprints.org> 
http://mailman.ecs.soton.ac.uk/mailman/listinfo/goal

_______________________________________________
GOAL mailing list
GOAL at eprints.org <mailto:GOAL at eprints.org> 
http://mailman.ecs.soton.ac.uk/mailman/listinfo/goal

_______________________________________________
GOAL mailing list
GOAL at eprints.org <mailto:GOAL at eprints.org> 
http://mailman.ecs.soton.ac.uk/mailman/listinfo/goal 

_______________________________________________
GOAL mailing list
GOAL at eprints.org <mailto:GOAL at eprints.org> 
http://mailman.ecs.soton.ac.uk/mailman/listinfo/goal 

_______________________________________________
GOAL mailing list
GOAL at eprints.org <mailto:GOAL at eprints.org> 
http://mailman.ecs.soton.ac.uk/mailman/listinfo/goal

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.ecs.soton.ac.uk/pipermail/goal/attachments/20140921/acb0feca/attachment-0001.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: image/png
Size: 925 bytes
Desc: not available
Url : http://mailman.ecs.soton.ac.uk/pipermail/goal/attachments/20140921/acb0feca/attachment-0001.png 