Proxmox VE 4.3 released!

reetp

Active Member
Aug 19, 2013
45
2
28
As it looks like you are decent UI/UX designer, please suggest a better layout, I am waiting for your suggestions. Please keep in mind to include all upcoming functionalities.
Here's a suggestion. Please get shot of the awful second vertical list. It is a complete and utter eyesore, and nothing is where you expect it. Put the menu back horizontally where it is expected. Fill in the big white blank space to the right of 'Summary' where it used to go.

Previous layout may not have been brilliant but it was 110% better and more usable than this one.

Earlier suggestion about a theme manager is very sensible - it would mean you could properly test themes and get some user feedback. Having this nonsense foisted on us with no choice is just irresponsible.

One seriously unhappy user.
 

dcsapak

Proxmox Staff Member
Staff member
Feb 1, 2016
4,019
365
88
31
Vienna
i want to chime in here (mainly because i implemented the most of the change to the vertical menu)

there were several reasons for why we wanted to do this: (from most important to least important (from user perspective))

  1. The Menu structure was not good. Mentally you had to change orientation up to 3 times, for example:

    You search your VM in the tree on the left (vertical),
    You search your Tab in the menu (horizontal, first orientation change), e.g. Firewall
    You click e.g. IPSet in the submenu (vertical, second orientation change)
    You look for the option you want, eg. create (horizontal, thrid orientation change)

    Now you have to

    Look for VM (vertical)
    Look for Firewall->IPSet (vertical, now not even a 2nd load)
    Look for create (horizontal, first orienation change)

    I find this much better, but i guess this can be subjective

  2. We want to add/group some Options to nodes, but
    we already had so many options, that even on a reasonably sized monitor,
    some where hidden (and maybe some people would never know they are there)

    also when using the extjs tabpanel layout, there were no options for groups,
    and it was very slow
  3. The Options were not very distinguishable

    what in mean by this is that there was only a single line of text, with only
    space in between, which made it hard to visually parse,
    now every option is on its own line

  4. We save vertical space

    most monitors nowadays are 16:9, so saving even 30 pixels vertical can make a difference

  5. We could remove much code

    this is more a technical reason, but without the ha/firewall/ceph submenus, we were able to remove this
    whole part, which was not so good, this is now much nicer (from a code standpoint)
There are however several points which are valid:
  • on 1920x1080 only 1 panel fits horizontally
    this is a bug and will be fixed, we generally optimize for 1280x720 and 1920x1080
  • too many options
    in the first step, we wanted to shift the menu to vertical
    the next step would be to group/hide some entries by default, but
    this (and which options) is still up to debate

Also, if someone has a better idea/design, you can always open a feature request on https://bugzilla.proxmox.com
or write to our developer list (in the best case you directly send patches)

but please, do not just post/write: "i do not like this, change it back" without constructive feedback
 

reetp

Active Member
Aug 19, 2013
45
2
28
i want to chime in here (mainly because i implemented the most of the change to the vertical menu)
Sorry to be so critical. It is not a 'personal' attack on you - merely an objective comment on the UI. I know as a designer it hurts when someone criticises all your hard work, but that's part of the job spec :) You cannot expect to be right every time.

there were several reasons for why we wanted to do this: (from most important to least important (from user perspective))

  1. The Menu structure was not good. Mentally you had to change orientation up to 3 times, for example:

    You search your VM in the tree on the left (vertical),
    You search your Tab in the menu (horizontal, first orientation change), e.g. Firewall
    You click e.g. IPSet in the submenu (vertical, second orientation change)
    You look for the option you want, eg. create (horizontal, thrid orientation change)

    Now you have to

    Look for VM (vertical)
    Look for Firewall->IPSet (vertical, now not even a 2nd load)
    Look for create (horizontal, first orienation change)

    I find this much better, but i guess this can be subjective
Yes it IS subjective - VERY. I never found the old way difficult. The new way just confuses....

Trying to be clever with GUI design is always a risk. I cannot think of anywhere else that uses a double left column layout, and with the width of screens there is absolutely no need. As I have no way to revert the changes I cannot now compare and contrast. Hence my request for a 'Theme Manager' so the USER can decide what they like and what they don't. There is no reason to just scrap the old and throw in the new - they could be run side by side.


  1. We want to add/group some Options to nodes, but
    we already had so many options, that even on a reasonably sized monitor,
    some where hidden (and maybe some people would never know they are there)

    also when using the extjs tabpanel layout, there were no options for groups,
    and it was very slow
  2. The Options were not very distinguishable

    what in mean by this is that there was only a single line of text, with only
    space in between, which made it hard to visually parse,
    now every option is on its own line
As above - I can't compare anymore.... I only know I was comfortable before, and confused now.


  1. We save vertical space

    most monitors nowadays are 16:9, so saving even 30 pixels vertical can make a difference
Sorry - if you have a wide screen then why try and cram everything to the left? It just looks awful and confused. If you want to save vertical space (which I have never found necessary on my wide screens) why not work out what to do with the logging window at the bottom which can take up a lot of space - why not have a simple button to open/close it ? A feature you see in a lot of places with a simple ^ v button to raise or lower the panel ???


  1. We could remove much code

    this is more a technical reason, but without the ha/firewall/ceph submenus, we were able to remove this
    whole part, which was not so good, this is now much nicer (from a code standpoint)
Removing code at the expense of usability is never a good idea, however tempting.

There are however several points which are valid:
  • too many options
    in the first step, we wanted to shift the menu to vertical
    the next step would be to group/hide some entries by default, but
    this (and which options) is still up to debate
Everyone else in the world seems to manage this with a horizontal bar and dropdowns. It is a standard that has ruled the world for years, and no one has found a better way to do it as yet - unless you can show me somewhere that has ?

Also, if someone has a better idea/design, you can always open a feature request on https://bugzilla.proxmox.com
or write to our developer list (in the best case you directly send patches)
I can open a bug but will it actually be listened too ? I'm not sure you think I have a point so I am unlikely to get anywhere. I'd love to try and hack some of it about or replace some of what I have before, but a) I'm not that good with JS and CSS and b) I can't see to find anywhere in your docs that leads me to where I should be looking to even try.

but please, do not just post/write: "i do not like this, change it back" without constructive feedback
What else do you want me to say ? For me the current layout is awful. The previous one was much better. two vertical columns just do not work, period. There is nothing much more 'constructive' that I can say right now... nothing you can do with the current layout will make it any better IMHO. Go back to the old, and start again. Sorry - all that hard work for nothing I know. But sometimes it's better to start again rather than flog a dead horse. Think Windows Metro......

A theme manager is the way to go - you can then add a new theme an check for customer feedback without forcing them to accept whatever it is that you consider to be great. Perhaps if you had given this option before foisting this theme on users you may have had some more comments.....

Have made some notes on a screenshot for you to consider.

Rgds
John
 

Attachments

  • Like
Reactions: inteam

reetp

Active Member
Aug 19, 2013
45
2
28
https://bugzilla.proxmox.com/show_bug.cgi?id=1137

Closed WONTFIX - you wonder why you bother.

The curious part is that the manager uses extjs which is a THEME MANAGER...... just that Prox don't want to allow you to swap themes for some unknown reason - certainly I haven't heard a sensible argument yet. If I could find the original files I am pretty sure this could be implemented without too much bother (might take me a while as I am useless with JS etc but I am happy to try).

Knowing all that I am pretty sure it would be a fairly trivial exercise for Proxmox devs, if they so chose, but my guess is there are other agendas at play.

Rgds
John
 

tom

Proxmox Staff Member
Staff member
Aug 29, 2006
13,816
448
103
Knowing all that I am pretty sure it would be a fairly trivial exercise for Proxmox devs, if they so chose, but my guess is there are other agendas at play.

Rgds
John
John, I really do NOT like your posting style. Telling the world that all this is a trivial task is just non-sense and our goals are hidden (other agenda?) is a bit weird.

Please come back to a friendly and constructive posting style and talk about facts and suggestions.
 

reetp

Active Member
Aug 19, 2013
45
2
28
Tom,

I'm sorry if you feel offended by me making a criticism. Such is the nature of development (I work on Koozali SME server so can speak from some experience). It appears from comments I am not alone.

Creating a bug that instantly gets marked Wontfix doesn't help, and why I was sceptical of trying as per your GUI designer. But I did and was promptly put down. Great.

I have said I think the second vertical menu is not good. I have provided an image and constructive suggestions.

I know that the menu system is controlled by extjs, the whole point of which is to enable you to have selectable themes (as it is a theme manager system), but you have chosen to disable that functionality and give users no choice which I do not understand. You could have enabled template selection, left the old template and made the new onecselectable and given users a choice. It is illogical, so logically there must be another reason behind your decision.

I asked where we could try developments such as this. No comment.

I asked where the files were to have a look myself. No comment.

I asked which commit made the changes. No comment.

And you wonder about my points ?

So far I have made a criticism, provided reasoning, provided suggestions, and your only reply is you don't like what I said.

Note I pay for Proxmox as well..... as if that makes a difference.

As a dev said to me once "Nothing personal. It's purely development"

So please, just answer my questions with logical arguments. I won't take it personally......

Rgds
John
 
  • Like
Reactions: inteam

tom

Proxmox Staff Member
Staff member
Aug 29, 2006
13,816
448
103
You got a very long answer to your bug report, so it was not just closed.

There are already improvements, e.g. for the 1920x1080 screen, so just follow our devel-mailing list if you want to be up2date to the very latest commits.
see also https://git.proxmox.com
 

Abergr

New Member
Oct 2, 2016
8
1
3
42
Im pretty new to Proxmox, still evaluating it, trying to figure out if we should invest time and money in this solution. Started a few weeks ago with 4.2, and just got the 4.3 update with the new UI, so I don't really have a habits yet, and yet I totally agree with you. I also think you have been very constructive, and I get the feeling that the people at Proxmox are not that humble towards their customers, which makes me hesitate a bit about this solution..

And to sum it up, the new UI is very confusing, don't get it at all. Really trying not to care, just to get work done, but keep struggling finding my way through the UI..
 
  • Like
Reactions: reetp

Abergr

New Member
Oct 2, 2016
8
1
3
42
Answering my own bug report, found out the issue since it happen again on another machine, on another vm.
I did a change of MAC address before the reboot, and there is apparently a bug if you using a odd number.
Just remove the Mac address of the network cards, so they are auto generated, and the container comes back to live.

Here you go:

arch: amd64

cpulimit: 8

cpuunits: 1024

hostname: vmlb01

memory: 2024

net0: name=eth0,bridge=vmbr1,hwaddr=3A:33:62:37:35:62,ip=192.168.0.205/24,type=veth

net1: name=eth1,bridge=vmbr0,gw=74.50.x.x,hwaddr=66:66:30:62:35:36,ip=74.50.x.x/28,type=veth

net2: bridge=vmbr2,hwaddr=36:30:37:36:65:30,ip=192.168.1.205/24,name=eth2,type=veth

onboot: 1

ostype: debian

rootfs: local-lvm:vm-100-disk-2,size=30G

swap: 51
 

reetp

Active Member
Aug 19, 2013
45
2
28
You got a very long answer to your bug report, so it was not just closed.
Sorry - please look again. https://bugzilla.proxmox.com/show_bug.cgi?id=1137

Resolved, WONTFIX. That's a closed bug if I am not mistaken. I posted a reply to Wolfgang and got zero response (as I expected)

There are already improvements, e.g. for the 1920x1080 screen, so just follow our devel-mailing list if you want to be up2date to the very latest commits.
see also https://git.proxmox.com
Yup - I'll subscribe. I like bashing my head against a brick wall :)

In the meantime could you please answer some of my questions above ? Seems I am not alone in my thinking from the looks of this thread.

Rgds
John
 

dcsapak

Proxmox Staff Member
Staff member
Feb 1, 2016
4,019
365
88
31
Vienna
i'll try to answer to you, so that we can clear everything up

Tom,

I'm sorry if you feel offended by me making a criticism. Such is the nature of development (I work on Koozali SME server so can speak from some experience). It appears from comments I am not alone.

Creating a bug that instantly gets marked Wontfix doesn't help, and why I was sceptical of trying as per your GUI designer. But I did and was promptly put down. Great.
yes, you opened a bug report, but you did the thing i explicitly asked you to do not, asking to change it back (or at least partially)
you then got a long response, in which it is explained, why it is not feasible for us to do this (in the near future)
also, a bug can (in general) be reopened

I have said I think the second vertical menu is not good. I have provided an image and constructive suggestions.
yes you provided an image, in which 2/3 arguments are already taken care of
the bottom bar, already has a triangle where you can minimize it, maybe you just missed it
and the empty space should be gone with an update in the near future
see https://git.proxmox.com/?p=pve-manager.git;a=commitdiff;h=155dce49acbfc9222d6f4567ef288147e0cc31f3

I know that the menu system is controlled by extjs, the whole point of which is to enable you to have selectable themes (as it is a theme manager system), but you have chosen to disable that functionality and give users no choice which I do not understand. You could have enabled template selection, left the old template and made the new onecselectable and given users a choice. It is illogical, so logically there must be another reason behind your decision.
i think you misunderstood what extjs is.
it is not a theme manager, but an javascript application framework.

we do not simply put buttons/menus in place, but there is quite a bit of logic inside the webgui
(e.g. the help button which opens a contextual help, depending on which suboption you have selected)

changing the color-theme is not that simple, because on numerous places, we had to manually adjust it, because
the extjs defaults are not very good, and with another theme, this work would multiply with every theme we have to support

changing the structure as a theme is even harder, because the logic is tied in with how the gui is generated

I asked where we could try developments such as this. No comment.

I asked where the files were to have a look myself. No comment.
the developer documentation is here:
https://pve.proxmox.com/wiki/Developer_Documentation

the git repositories are here:
https://git.proxmox.com

and the files for the webgui are here (the www/manager6 folder in the pve-manager repo)
https://git.proxmox.com/?p=pve-manager.git;a=tree;f=www/manager6;h=234ea4bb5721ab1b3696944546a82f4fc59b856c;hb=refs/heads/master

I asked which commit made the changes. No comment.
ok this is a hard one, because there were many changes which were involved in this, but I'll try:
https://git.proxmox.com/?p=pve-manager.git;a=commitdiff;h=ae21ba1c447fbedde08ed5668dae8e8b095738bc
https://git.proxmox.com/?p=pve-manager.git;a=commitdiff;h=4ad7907769b80f68531bc3f4e7133d753c3efdd5
https://git.proxmox.com/?p=pve-manager.git;a=commitdiff;h=5a0dc87f85c38011f3f2e4a91c2eda4c5f30d194
https://git.proxmox.com/?p=pve-manager.git;a=commitdiff;h=54069425e21ce478542c1a951fc327b493507e7d
https://git.proxmox.com/?p=pve-manager.git;a=commitdiff;h=76682e0290b81157001bff95506eb7f84e54ee91
https://git.proxmox.com/?p=pve-manager.git;a=commitdiff;h=90a50103da0d4eaf1824479c1bc2b514fd35fd24
https://git.proxmox.com/?p=pve-manager.git;a=commitdiff;h=172978d27d407f036bd37cb894f87f6b4a66e595
https://git.proxmox.com/?p=pve-manager.git;a=commitdiff;h=958049cb52441f57fa92492161a0d4e97bee8287
https://git.proxmox.com/?p=pve-manager.git;a=commitdiff;h=3ee430ea182b976b049c781d2a0e99dee1987f9c
https://git.proxmox.com/?p=pve-manager.git;a=commitdiff;h=0c43baa8e088b71f54b43b6e706b4352cf860cdf

and every commit which touches on of the files changed above would have to be looked at (but there are too many to list them here; also i probably forgot one or two commits where there were changes for this)

And you wonder about my points ?

So far I have made a criticism, provided reasoning, provided suggestions, and your only reply is you don't like what I said.

Note I pay for Proxmox as well..... as if that makes a difference.

As a dev said to me once "Nothing personal. It's purely development"

So please, just answer my questions with logical arguments. I won't take it personally......

Rgds
John
yes, you provided criticism, but the suggestions are sadly not so easily implemented and impose a high impact on what we can develop and on the maintenance of the code in the future (our time is sadly limited)

i hope you can understand this, and if you decide to learn how to implement something you want, you send us a patch for it
 

informant

Member
Jan 31, 2012
674
6
18
could you post the output of "pveversion -v"?
Hi yes:
pveversion -v
proxmox-ve: 4.3-66 (running kernel: 4.2.2-1-pve)
pve-manager: 4.3-1 (running version: 4.3-1/e7cdc165)
pve-kernel-4.4.13-1-pve: 4.4.13-56
pve-kernel-4.4.13-2-pve: 4.4.13-58
pve-kernel-4.4.16-1-pve: 4.4.16-64
pve-kernel-4.4.19-1-pve: 4.4.19-66
lvm2: 2.02.116-pve3
corosync-pve: 2.4.0-1
libqb0: 1.0-1
pve-cluster: 4.0-46
qemu-server: 4.0-88
pve-firmware: 1.1-9
libpve-common-perl: 4.0-73
libpve-access-control: 4.0-19
libpve-storage-perl: 4.0-61
pve-libspice-server1: 0.12.8-1
vncterm: 1.2-1
pve-qemu-kvm: 2.6.1-6
pve-container: 1.0-75
pve-firewall: 2.0-29
pve-ha-manager: 1.0-35
ksm-control-daemon: 1.2-1
glusterfs-client: 3.5.2-2+deb8u2
lxc-pve: 2.0.4-1
lxcfs: 2.0.3-pve1
criu: 1.6.0-1
novnc-pve: 0.5-8
fence-agents-pve: not correctly installed
 

fabian

Proxmox Staff Member
Staff member
Jan 7, 2016
3,527
546
118
hi, after upgrade many errors:
Sep 30 10:56:19 andromeda systemd-sysv-generator[10761]: Ignoring creation of an alias umountiscsi.service for itself
Sep 30 10:56:19 andromeda systemd[1]: [/run/systemd/generator.late/qemu-server.service:7] Failed to add dependency on +iscsi.service, ignoring: Invalid argument
Sep 30 10:56:19 andromeda systemd[1]: [/run/systemd/generator.late/rgmanager.service:7] Failed to add dependency on +vz.service, ignoring: Invalid argument
Sep 30 10:56:19 andromeda systemd[1]: [/run/systemd/generator.late/rgmanager.service:7] Failed to add dependency on +qemu-server.service, ignoring: Invalid argument
the first is a bug in jessie's openiscsi, but harmless. the rest seems like leftover cruft from a 3.x installation - is this an upgraded node? if you have the files "/etc/init.d/qemu-server" and/or "/etc/init.d/rgmanager", you can remove them.

and by click on disc it comes "no disc found" we use hp smart array in raid...
this should be fixed soon.
 

reetp

Active Member
Aug 19, 2013
45
2
28
i'll try to answer to you, so that we can clear everything up
yes, you opened a bug report, but you did the thing i explicitly asked you to do not, asking to change it back (or at least partially)
you then got a long response, in which it is explained, why it is not feasible for us to do this (in the near future)
also, a bug can (in general) be reopened
Sorry - so what are bug reports for if not for raising issues ?

Also, if someone has a better idea/design, you can always open a feature request
You said nothing 'explicit' about 'don't ask to change it back' as far as I can see. In any case, if you bother to read VERY carefully you will see that I didn't.

The bug was entitled "Add theme manager" and yes, I chose my words there VERY carefully, on purpose. Because I knew if I just said 'change it back' I would have got two fingers. So, I asked you to add a theme manager as an enhancement. I asked where the files were so *I* could look at changing them back.

Please go and read again what I actually said https://bugzilla.proxmox.com/show_bug.cgi?id=1137 before making accusations which *I* find offensive.

yes you provided an image, in which 2/3 arguments are already taken care of
the bottom bar, already has a triangle where you can minimize it, maybe you just missed it
Yup - it is so tiny that I did miss it.... I thought that was only for resizing by click/drag....... 2/3rds taken care of ? Hmmmmmm.....

In that case even less reason to have 2 vertical menus. If you removed the vertical menu and added it horizontally back in to the empty space as I pointed out, then you'd have more room and wouldn't need this patch. Talk about making work for yourselves.

i think you misunderstood what extjs is.
it is not a theme manager, but an javascript application framework.

we do not simply put buttons/menus in place, but there is quite a bit of logic inside the webgui
(e.g. the help button which opens a contextual help, depending on which suboption you have selected)

changing the color-theme is not that simple, because on numerous places, we had to manually adjust it, because
the extjs defaults are not very good, and with another theme, this work would multiply with every theme we have to support

changing the structure as a theme is even harder, because the logic is tied in with how the gui is generated
Yes I understand it is a framework, but it has the innate ability to handle themes. Why ignore that functionality ?

As I have previously said, I am no guru at this. However I can see from the demo site at extjs you can have multiple themes. So why can't Promox, as I suggested ?

http://docs.sencha.com/extjs/6.2.0/guides/core_concepts/theming.html
http://examples.sencha.com/extjs/6.2.0/examples/classic/themes/index.html (note the selectable themes)

Seems to work OK ? So why not add a theme selector as per the page and then you can add in your new theme and let people choose ?

I think the code you needed to add was probably very simple :

http://extjs.eu/lightweight-theming/

the developer documentation is here:
https://pve.proxmox.com/wiki/Developer_Documentation

the git repositories are here:
https://git.proxmox.com
Yup, I got those.

and the files for the webgui are here (the www/manager6 folder in the pve-manager repo)
https://git.proxmox.com/?p=pve-manager.git;a=tree;f=www/manager6;h=234ea4bb5721ab1b3696944546a82f4fc59b856c;hb=refs/heads/master

ok this is a hard one, because there were many changes which were involved in this, but I'll try:
Thank you. I can take a look now.

yes, you provided criticism, but the suggestions are sadly not so easily implemented and impose a high impact on what we can develop and on the maintenance of the code in the future (our time is sadly limited)
I have criticised with valid reasons for the criticism, and tried to provide some positive suggestions for change. As I develop a bit myself I know that this is the way to do it. As a developer you have to be able to handle that, and to accept when you are point blank wrong, no matter how great your idea may be.

i hope you can understand this, and if you decide to learn how to implement something you want, you send us a patch for it

Actually I don't quite. You had to do a lot of work to change the theme that hasn't exactly been completely popular. It may have been beneficial to have implemented a theme manager as a logical first step so that you could then test new themes. You may have gotten more feedback that way and before diving headlong down the rabbit hole with no route back.

As far as it goes, it probably wouldn't have been too difficult to add the old template back if you so chose. You have all the code there already so it isn't like you have to write anything new. If, as you suggest, the new template changes lots of files beyond the core template, one has to question what is the point of using a framework/CSS etc - the point was you separate style from content surely - so you should be able to change a style/template without breaking anything ? Or I have I missed something in web app design over the last 20 years ?)

You could have add a theme selector widget, added your new theme and then allowed a choice and ask for comments. That would have been a much more sensible decision.

However, I will have a look for myself on how to get out of this because it grates more every time I look at it. If I can figure how to add a theme selector I'll post the code.

FYI I have applied to the dev list yesterday but so far heard nothing - I assume it needs moderator approval. I'd be pleased if you could get that followed up.

Sorry but you may sense my frustration at not seeing a single sensible argument to support your position. No one uses double vertical menus - I've tried to find an example, but so far can't. There is a very good reason for that. It just doesn't work, and certainly doesn't in your implementation. If it did, everybody would be using it. But you have chosen to use it, and despite suggestions that it may not have been such a good move, you clearly have no intentions of considering other options.

I've tried my best to make positive suggestions and illustrate the folly of your ways, but there just seems no room for discussion. I therefore feel like I am hitting my head against brick wall, and that is very frustrating.

And that makes me think about whether I will bother to renew my subscription, and I hope others think twice about it as well.

Rgds
John
 

t.lamprecht

Proxmox Staff Member
Staff member
Jul 28, 2015
1,719
260
103
South Tyrol/Italy
Sorry - so what are bug reports for if not for raising issues ?
personal taste != issue, and constructive critic is something else, IMHO. You merely rant and scold at PVE, mixed with some not bad critic, but that got already addressed by us in the last commits. So its not that we did not value your posts.

I mean for me it seems like previously you used PVE and probably even liked it, or at least didn't hated it. Now this - relative to the whole update with under the hood changes, fixes, features ... - small change lets you make it unusable and make it the worst thing ever.
I really cannot relate to such behaviour, sorry, we normally really welcome feedback and implement features and have often great discussions, the community has their voice, since I'm at Proxmox at least half the stuff I made was for that reason.

You said nothing 'explicit' about 'don't ask to change it back' as far as I can see. In any case, if you bother to read VERY carefully you will see that I didn't.
but please, do not just post/write: "i do not like this, change it back" without constructive feedback
... its here...

In that case even less reason to have 2 vertical menus. If you removed the vertical menu and added it horizontally back in to the empty space as I pointed out, then you'd have more room and wouldn't need this patch. Talk about making work for yourselves.
You're quite patronizing here, or do I misinterpret this? I mean d.csapak took quite some time being understanding and to explain the reasons, that besides usability it had technical reasons (big code cleanup, which means less maintaining work in the future), and you just talk him down like if he is/ we are stupid and if we just followed your personal taste everything would be better...

I have nothing to do with the vertical menu layout and was in fact confronted by it when I cam back from vacation.
First naturally its a change so I needed a few minutes to accommodate my navigation, but for me it was an improvement after that.

My scenarios are always:
Search something in the left Resource Tree, so I have to go down vertically on that list with my eyes, when I found it I can switch over a centimeter with my eyes and have all the options in place, no need to move my mouse far, no need to switch orientation, I can get to the changes I want to make faster.
The top is no reserved for panel specific stuff, edit/remove/... buttons for me it seems cleaner
That said, its surely not perfect!

And as he said, what if more options come, whats then better, a vertically grouped menu or a long list of tab buttons you have to search for?
And regarding drop downs, they are a good way to hide options and make the user search for them, further they vanish the half time or if you make the persistent on click they are in the way half the time and overlap content.
Being as objective as I can i wouldn't use them for PVE, and as a fact personally I wouldn't too, but that doesn't really matter here.

Investing weeks in providing ten other themes and then another few weeks per year maintaining those seems to me like time better invested in real problems, that's the real point we try to make.
Just dirty hacking in another theme once is surely doable :)

Sorry but you may sense my frustration at not seeing a single sensible argument to support your position.
I've tried my best to make positive suggestions and illustrate the folly of your ways, but there just seems no room for discussion. I therefore feel like I am hitting my head against brick wall, and that is very frustrating.
I can relate to your frustration, but you maybe also sense that we feel a bit the same here, so maybe you can relate to ours here too.

I've tried to find an example, but so far can't.
2016-10-07-090442_779x638_scrot.png



I didn't need to search long :) Against FreeNAS 10 we're quite conservative, but I guess they more configurable your product is the more you have problems with displaying the options in a sane way :)

And that makes me think about whether I will bother to renew my subscription, and I hope others think twice about it as well.
I hope you can still use Proxmox VE for the good or find a better alternative which fits your graphical needs more. And I hope others are not intimidated by this "brick wall discussion".

I do not know if it makes sense do continue this, our points were made, you decided they were invalid or not reasonable, we tried to explain why we did it this way and did not heard a better alternative yet, at least not one that can be done without adding weeks of work per year for not real value added. Repeating the same over and over won't help anyone.
 

Rhinox

Active Member
Sep 28, 2016
272
36
28
29
This whole discussion is just another example of "we know what is better for you" attitude. I'm not sure it is the right way to go...
 
  • Like
Reactions: m.ardito and reetp

gkovacs

Well-Known Member
Dec 22, 2008
503
45
48
Budapest, Hungary
And I hope others are not intimidated by this "brick wall discussion"..
This whole discussion is just another example of "we know what is better for you" attitude. I'm not sure it is the right way to go...
I think there is a more important issue here than the new navigation, which looks a bit unconventional and takes some getting used to, but I personally think is actually better from a usability standpoint (your eyes need to scan a much smaller area to find the next click). What I personally don't like about the new (version 4) GUI is the font (and list) size increase compared to the previous version, hopefully we will get an option to use smaller fonts in the future. But I digress.

The more important issue is governance of an open source project: namely, who decides what new functions get implemented, and how. Proxmox is developed by a small team working in the same office (I presume), with help from a handful of contributors, but it is used by a huge number of users, probably in the tens (or hundreds) of thousands. Some discussions about the roadmap (new feature requests and suggestions) usually happen on the forums, some even get to the bugzilla, but the decision about them is not transparent most of the time, and happens in the Proxmox offices or on the developer mailing list (again I presume).

As the project matures and the number of users increases, this needs to change into a more organized, more involving, more transparent system. Especially since there are thousands of users who pay for some kind of support license now, their voices need to be heard, but regular users as well. At this very moment I don't have any particular suggestion how to move forward, but I think it's time the Proxmox developers take a hard look about that roadmap process, and hopefully implement some kind of system that helps them decide how to allocate those finite resources with a bit more input from the userbase.

I would very much like to vote on new features for example.
 
Last edited:
  • Like
Reactions: m.ardito

t.lamprecht

Proxmox Staff Member
Staff member
Jul 28, 2015
1,719
260
103
South Tyrol/Italy
I think there is a more important issue here than the new navigation, which I admit looks a bit unconventional and takes some getting used to, but I personally think is actually better from a usability standpoint (your eyes need to scan a much smaller area to find the next click). What I personally don't like about the new (version 4) GUI is the font (and list) size increase compared to the previous version, hopefully we will get an option to use smaller fonts in the future. But I digress.
The font size change happened from 3.4 to 4.X new theme? Or did my eyes just oversaw a new Change from 4.2 to 4.3?

The more important issue is governance of an open source project: namely, who decides what new functions get implemented, and how. Proxmox is developed by a small team working in the same office (I presume), with help from a handful of contributors, but it is used by a huge number of users, probably in the tens (or hundreds) of thousands. Some discussions about the roadmap (new feature requests and suggestions) usually happen on the forums, some even get to the bugzilla, but the decision about them is not transparent most of the time, and happens in the Proxmox offices or on the developer mailing list (I presume).
Thank you for your well formed post and constructive critic!

Your presumes are correct as far as I can see.
The only thing I would say that a lot of decisions happen at the pve-devel mailing list, but I fully understand that this list is a) not small in volume, things can get lost b) is for developers so for a lot of users it is invisible.
I also get often my first feedback on patches there from a colleague and or contributor, but yes we plan/talk about some stuff in our small office, naturally :)

As the project matures and the number of users increases, this needs to change into a more organized, more involving system. Especially since there are thousands of users who pay for some kind of support license now, their voices need to be heard, but regular users as well. At this very moment I don't have any particular suggestion how to move forward, but I think it's time the Proxmox developers take a hard look about that roadmap process, and hopefully implement some kind of system that helps them decide how to allocate those finite resources with input from the userbase.
I can relate to that, we thought a bit of that here also (but remember this is my personal opinion I'm posting here :) ) I and a colleague had the Idea of blog post (or a similar medium), which happen semi frequently (e.g. 1 - 3 a month or something like that) which may show some screenshots on a GUI change, or a short paragraphs about stuff we're doing currently (even if its just experimenting).
Maybe I can also start this as a private one, as I'm able to follow the pve-devel list (unsurprisingly :) ) so I could setup a small blog on my site and try this as an experiment.

But the idea has also problems, often we can show stuff only when its almost finished where the "planning" phase is almost over and its also simply work which has to be done. Also it does not guarantee that a change does not happen, I mean general the new menu layout was not taken bad, a percentage had real problems with it and you never will make everyone happy, you can do what your want, each projects has is advantages and drawbacks.

I would like to see additional involvement from the community, especially in development, a lot of you do already amazing stuff with helping others here in the forum, testing, ... thanks for that!
Even if you "just" setup a old machine or a VM and configure Proxmox VE there with the pve-test repository, you then often see things much earlier and can give (objective) feedback or help us fix bugs earlier.
And naturally contributing is also an option. This is the way most Open Source projects handle decision. The main developers do not have time for a wished feature or change, or they feel it isn't that important: you really want this in our project, then develop and contribute it and make it to your project as well. I know that's not an option for all.
But what see here my colleagues and I try often accommodate to feedback, but sometimes i have to say no to something which i do not see good in PVE or which i cannot bring up the time to develop and maintain it for the value it brings.

Note again, this lengthy post is my personal opinion.
 

hansm

Member
Feb 27, 2015
61
3
8
I just want to say the new menu works fine for us. A product evolves and changes are necessary sometimes, you just have to adjust to it. Doesn't happen this all the time in software, and the rest of the world? Your new microwave probably doesn't use the same controls as your old but it reallly does the same thing, you only need to adjust to this new control. Really most software I know is changed in look-and-feel over the years for whatever reason. Double vertical menus are also used in some places in VMware Vsphere Client and Magento 2 uses some kind of double vertical menu.

(big code cleanup, which means less maintaining work in the future)
Wouldn't this mean the developers have more time to work on feature requests and bugfixes reported by the community?? Good news!

This whole discussion is just another example of "we know what is better for you" attitude.
I don't think so. The Proxmox team seems to be busy working on their product for us. They develop new features and give us a wonderful product. They don't need our approval for every change they want to make, you can like or hate something new, well, you can tell the community about it and if a large group shares that opinion, Proxmox will reconsider their changes and probably undo them. They are also implementing feautures wanted by the community, in my opinion they listen to the community and communicate with us all the time at the forums.

I really think the Proxmox team is doing a great job and they explained very well why they changed the menu, I understand their reasons and they are very valid in my opinion. Spending less time maintaining the code and prepare the product for even more features! Why would you be against it...

Regards from a happy user/customer. Keep on the great work!
 
  • Like
Reactions: wosp

About

The Proxmox community has been around for many years and offers help and support for Proxmox VE and Proxmox Mail Gateway. We think our community is one of the best thanks to people like you!

Get your subscription!

The Proxmox team works very hard to make sure you are running the best software and getting stable updates and security enhancements, as well as quick enterprise support. Tens of thousands of happy customers have a Proxmox subscription. Get your own in 60 seconds.

Buy now!