Lucas Taylor, CMS Communications Meeting 1April 18, 2023

Lucas Taylor, CMS Communications Meeting 2April 18, 2023

Video-conferencing tools choice: EVO or Vidyo ?

Working model, already agreed by expts and CERN- Expts define needs, CERN/IT provides services, expts pay (most) bills

EVO meets most of CMS needs so why consider other options?- Continually seek to improve communications in CMS and minimise costs- Major external procurements need to consider all suitable suppliers

CERN assessed the market, finding two strong candidates- EVO: > 80 % of usage is LHC, HEP developed (~5 FTE) - Vidyo: many users. Licensed by Google, Cisco (~ 100 FTE, commercial)

CERN/IT (+ expts) will now assess the tools in terms of - Do tools meet CMS needs? Functionality, quality, scale, ease of use…- Are the risks acceptable? Technical, commercial, etc.- What is the total cost of ownership? Licence fees, support, etc.

CMS shall take an active role in this assessment/decision

LT Draft Proposal for EVO / Vidyo Assessment Process

Is Vidyo a candidate to meet

CMS needs ? EVO in Production

Migrate to Vidyo

No

Does Vidyo production service

meet CMS needs?

Which system is better ?

…functionality, costs, risks…

No

Yes

Yes

Aug 2010 Mar 2011

Establish fullVidyo service

Vidyo pilot service

EVO in Production

Vidyo in Production(trial service)

Oct 2010

Vidyo

Jun 2011

•Bridge Vidyo to existing systems as an additional option to EVO

•Start with test room (354 / 1-016) and test group (Computing Ops)

•Extend to all CMS meeting rooms at CERN and exercise by all CMS

Vidyo in Production

EVO

•Need to pay for both EVO and Vidyo for overlap period

•CERN / IT is exploring the costs

• If Vidyo is chosen, may re-coup costs (nature of licensing model)

•Need broad and clear consensus

•Need broad and clear consensus

A CMS for CMS ?Preliminary views on Web Content Management Systems

Lucas Taylor, FermilabCMS Head of Communications

(Web) Content Management Systems

DG Communications group has identified a number of general problems with CERN Web sites - No coherent architecture to domain: cern.ch- Poor navigation and search does not work- CERN does not have a website, too many top-level (entry) pages- Many problems – bad URLs, data duplication, unmanaged “design”

(as distinct from content)

DG / CO recommends (CERN Strategic Communication plan) the site-wide adoption of a Content Management System (CMS) which enables you to - Publish in one place, reuse in many- Manage design separately from content- Navigate, filter, version, support multiple languages - Workflows (create content, approve, publish, revise, revoke, etc.)

A lot of resonance with issues we face in CMSCMS Web Systems

Internal Web pages and services- Organisation / projects: people, management,

institutes, plans, schedules, resources, secretariat …- Communications: Email lists, news, blogs,

meetings, videoconf. …- Documents & Publications systems- CMS operations: e-logs, user support,

monitoring, SW dev. …

External Web pages and services- Scientists: papers, notes, contacts …- The public/press: physics, photos, movies,

educational resources …

Main Technologies

Linux, Mac, (Windows)

Apache, Tomcat

Firefox, Safari, (others)

html, css, PHP, Java, Python

Twiki, Emacs, Dreamweaver

Oracle, mySQL

e-groups, Hypernews

Indico, EVO

CDS, DocDB, EDMS

Twitter, YouTube

… and more …

Slides presented at brain-storming discussion with CERN / IT, CERN

Depts, LHC experiments, 18 May 2010

CMS Web Systems

Internal Web pages and services- Organisation / projects: people, management,

institutes, plans, schedules, resources, secretariat …- Communications: Email lists, news, blogs,

meetings, videoconf. …- Documents & Publications systems- CMS operations: e-logs, user support,

monitoring, SW dev. …

External Web pages and services- Scientists: papers, notes, contacts …- The public/press: physics, photos, movies,

educational resources …

Main Technologies

Linux, Mac, (Windows)

Apache, Tomcat

Firefox, Safari, (others)

html, css, PHP, Java, Python

Twiki, Emacs, Dreamweaver

Oracle, mySQL

e-groups, Hypernews

Indico, EVO

CDS, DocDB, EDMS

Twitter, YouTube

… and more …

Slides presented at brain-storming discussion with CERN / IT, CERN

Depts, LHC experiments, 18 May 2010

Issue 1: Muddled CMS Web entry point(s)

Too many CMS “Home Pages”- With sub-optimal design & content

Working on single CMS entry point branching out to- Public Web site- Collaboration Web site

Design, navigation and content still need a lot of work

Slides presented at brain-storming discussion with CERN / IT, CERN

Depts, LHC experiments, 18 May 2010

Issue 1: Muddled CMS Web entry point(s)

Too many CMS “Home Pages”- With sub-optimal design & content

Working on single CMS entry point branching out to- Public Web site- Collaboration Web site

Design, navigation and content still need a lot of work

Slides presented at brain-storming discussion with CERN / IT, CERN

Depts, LHC experiments, 18 May 2010

Issue 2: Too many sites. Poorly maintained.

CMS is large, complex, distributed, diverse, and hard to manage

There are 245 (!!) “official” CMS Web sites at CERN

- Plus the Twiki, and an unknown number of non-CERN sites

CMS Web entropy is ever increasing - It is (too!) easy to create new Web pages / sites- Maintenance is boring, responsibilities are ill-defined

Users do not have a culture of expecting good quality Web services. Resources are not made available.Slides presented at brain-storming

discussion with CERN / IT, CERN Depts, LHC experiments, 18 May 2010

Issue 2: Too many sites. Poorly maintained.

CMS is large, complex, distributed, diverse, and hard to manage

There are 245 (!!) “official” CMS Web sites at CERN

- Plus the Twiki, and an unknown number of non-CERN sites

CMS Web entropy is ever increasing - It is (too!) easy to create new Web pages / sites- Maintenance is boring, responsibilities are ill-defined

Users do not have a culture of expecting good quality Web services. Resources are not made available.Slides presented at brain-storming

discussion with CERN / IT, CERN Depts, LHC experiments, 18 May 2010

Issue 3: Incoherent tools, style, navigation

No coherent choice of hosts - Twiki, afs, Nice, sharepoint …

No coherent style - > 245 personal tastes

Navigation is miserable

There have been recent attempts to standardize more (header, sidebar, css)- Well-motivated but still not enough

(does not cover 244 of the 245 sites)- Should we adopt a full Content

Management System ?Slides presented at brain-storming discussion with CERN / IT, CERN

Depts, LHC experiments, 18 May 2010

Issue 3: Incoherent tools, style, navigation

No coherent choice of hosts - Twiki, afs, Nice, sharepoint …

No coherent style - > 245 personal tastes

Navigation is miserable

There have been recent attempts to standardize more (header, sidebar, css)- Well-motivated but still not enough

(does not cover 244 of the 245 sites)- Should we adopt a full Content

Management System ?Slides presented at brain-storming discussion with CERN / IT, CERN

Depts, LHC experiments, 18 May 2010

Issue 4: Many documents are not managed at all

Estimate ~ 100k CMS documents so far, many informal but containing valuable knowledge - ~50% already in iCMS, Indico, CDS,

EDMS- ~50% scattered about on various

Web sites, private disks, etc.

CMS recently started using DocDBto harvest these documents- Fermilab product- Running on CERN / IT systems- Will integrate/migrate to iCMS + CDS

in longer term

Estimate total CMS by extrapolation from Pixel

group

Slides presented at brain-storming discussion with CERN / IT, CERN

Depts, LHC experiments, 18 May 2010

Issue 4: Many documents are not managed at all

Estimate ~ 100k CMS documents so far, many informal but containing valuable knowledge - ~50% already in iCMS, Indico, CDS,

EDMS- ~50% scattered about on various

Web sites, private disks, etc.

CMS recently started using DocDBto harvest these documents- Fermilab product- Running on CERN / IT systems- Will integrate/migrate to iCMS + CDS

in longer term

Estimate total CMS by extrapolation from Pixel

group

Slides presented at brain-storming discussion with CERN / IT, CERN

Depts, LHC experiments, 18 May 2010

Issue 5: Hard to find (correct) information

… as a result of issues 1-4 …

It is very hard to find information- CMS Web lacks a well-designed (navigable) structure- No coherent search function for the many Web sites - Even Google often fails – many pages are protected

Much information of importance is duplicated, incomplete, out-of-date or plain wrong- People often create new pages (esp. on the Twiki) because they

cannot find existing ones or are not able to fix them - Then the new pages slowly decay, being neither updated nor deleted

As a result, CMS has poor access to its own knowledge base - This leads to inefficiencies, reduced competitiveness, or even errors - Longer term, we risk losing crucial CMS knowledge

Slides presented at brain-storming discussion with CERN / IT, CERN

Depts, LHC experiments, 18 May 2010

Issue 5: Hard to find (correct) information

… as a result of issues 1-4 …

It is very hard to find information- CMS Web lacks a well-designed (navigable) structure- No coherent search function for the many Web sites - Even Google often fails – many pages are protected

Much information of importance is duplicated, incomplete, out-of-date or plain wrong- People often create new pages (esp. on the Twiki) because they

cannot find existing ones or are not able to fix them - Then the new pages slowly decay, being neither updated nor deleted

As a result, CMS has poor access to its own knowledge base - This leads to inefficiencies, reduced competitiveness, or even errors - Longer term, we risk losing crucial CMS knowledge

Slides presented at brain-storming discussion with CERN / IT, CERN

Depts, LHC experiments, 18 May 2010

Next Step – Review CMS Web systems (June 2010)

Outcome: written requirements and a strategy for all CMS Web systems, including recommendations of key technologies

Membership: up to about a dozen people including - CMS Head of Communications (Chair)- Significant CMS “customers”: Management, Physics Groups,

Publications, Computing/Offline, Secretariat, Outreach- CMS experts: iCMS developers, Webtools …- External experts: e.g. CERN/IT, FNAL, ATLAS …

Modus Operandi: dynamic sub-groups, e.g. “Content”, “Design”, “Technologies”, etc.

Review should explicitly address the potential role of Content Management Systems

Slides presented at brain-storming discussion with CERN / IT, CERN

Depts, LHC experiments, 18 May 2010

Next Step – Review CMS Web systems (June 2010)

Outcome: written requirements and a strategy for all CMS Web systems, including recommendations of key technologies

Membership: up to about a dozen people including - CMS Head of Communications (Chair)- Significant CMS “customers”: Management, Physics Groups,

Publications, Computing/Offline, Secretariat, Outreach- CMS experts: iCMS developers, Webtools …- External experts: e.g. CERN/IT, FNAL, ATLAS …

Modus Operandi: dynamic sub-groups, e.g. “Content”, “Design”, “Technologies”, etc.

Review should explicitly address the potential role of Content Management Systems

Slides presented at brain-storming discussion with CERN / IT, CERN

Depts, LHC experiments, 18 May 2010

Joint Meeting of CERN Departments and the LHC experiments, 18 May 2010

CERN / IT – David Foster (IT Deputy Dept. Head) … leading the discussion plus Christian Isnard, Tim Smith, Andreas Wagner

CERN / DG Communications Group – Dan Noyes CERN / GS Dept. – David Widegren CERN / HR Dept – Catharina Hoch CERN / KTT – Manuela Cirilli (formerly ATLAS) ATLAS – Shaun Roe, Steven Goldfarb, Kathy Pommes, Joao

Pequenao CMS – David Barney, Marzena Lapka, Gilles Raymond, Lucas

Taylor

http://indico.cern.ch/conferenceDisplay.py?confId=95230 Group was given a name: ENTICE (Enterprise Needs for Tools and

Infrastructure for Content Exploitation)

Some general points arising from the discussion

Common solution, architecture and services are desirable Connectors are required to all existing repositories (Indico,

CDS, EDMS, etc.) It may be now “too easy” to create websites and publish

content - What is really official information is unclear

The management of content and its tagging etc. must be the responsibility of the content providers

A clear message is that open-source solutions are mature and credible with major public websites based on them- Based on well-known “LAMP stack” (Linux, Apache, MySQL,

Perl/Python/PHP)- Sharepoint usage is growing at CERN but is still perceived as

difficult to go beyond the “out of the box” options, finding external expertise is hard, considered as windows centric and expensive.

Towards a “CERN/LHC World Wide Web Strategy”LT asked to propose a CERN/LHC-wide framework to joint CERN/expts meeting on 9th June:

1. Questionnaire. Goal: Prepare a top-level summary of needs and issues- Participation: one contact person per CERN Dept. and experiment - Format: questionnaire to establish purpose of Web(s), formats, technologies,

issues faced (e.g. top five items). Tool to measure no. sites, pages, users, etc. ? - Output: standard summary table (~1 page) per Dept. / experiment

2. 1-day Workshop 1: Goal: identify common goals, issues, potential solutions- Participation: few people per Dept. / Expt. in dynamic working groups- Format: short sessions per topic, focused (provocative) discussions of goals, issues

and solutions, incl. potential role of a Content Management System- Output: concise written summary (< 10 pages, rapidly produced, not polished)

3. Broad consultation in CERN Depts. and Expts. Collect external advice.

4. 1-day Workshop 2. Goal: Draft “CERN/LHC World Wide Web Strategy” - Participation: a few people per Dept. / Expt. Plus a few external experts.- Format: Plenary. Discuss sections of a pre-agreed skeleton plan. Sketch out the

content/conclusions. Identify people to write text within < 1 week after meeting. - Output: (consensus) recommendations to CERN and experiment managements,

including potential role of a CERN-wide Web Content Management System

5. Decision by CERN and experiment managements

June 2010Sept

AugustJuly

DocumentManagement

System

Information (Web) Systems

Needs, Reviews, Decisions

Jan

– Ju

n, 2

010

(wor

k al

read

y in

pro

gres

s)20

11

Adopt a CMS ?

(Content Mgmt System)

Select / deployCERN-wide

CMS system

Address pressingneeds

Priorities

Lower

Address need to improve Communications efficiency (e.g. Email, Calendar, blogs,

meetings, forums, etc.)

Need to clean up

information Systems (Web)

Make new physics pages (iCMS / Twiki)

Review / define Web

Strategy (CERN-wide ?)

Set up interim document mgmt system (DocDB)

Need new physics results pages

Need to collect and secure ~100k

existing documents

Migrate documents to coherent CERN system (e.g. CDS-

based)

Assess situation

Identify issues

1) Home page(s) muddled2) Too many Web sites (245). Old. 3) Incoherent tools, navigation, style4) Many documents not in system5) Hard to find (correct) information

Harvest documents

from all sub-systems

Jul –

Dec

20

10

Hig

h

No

Yes

Clean up Web sites

(using current

systems)

Clean up Web sites

(using a CMS)

CMS Information Systems – Proposed Roadmap

DocumentManagement

System

Information (Web) Systems

Needs real resources (people): CERN/IT, CMS Communications

Group, CMS sub-systems

Needs, Reviews, Decisions

Jan

– Ju

n, 2

010

(wor

k al

read

y in

pro

gres

s)20

11

Adopt a CMS ?

(Content Mgmt System)

Select / deployCERN-wide

CMS system

Address pressingneeds

Priorities

Lower

Address need to improve Communications efficiency (e.g. Email, Calendar, blogs,

meetings, forums, etc.)

Need to clean up

information Systems (Web)

Make new physics pages (iCMS / Twiki)

Review / define Web

Strategy (CERN-wide ?)

Set up interim document mgmt system (DocDB)

Need new physics results pages

Need to collect and secure ~100k

existing documents

Migrate documents to coherent CERN system (e.g. CDS-

based)

Assess situation

Identify issues

1) Home page(s) muddled2) Too many Web sites (245). Old. 3) Incoherent tools, navigation, style4) Many documents not in system5) Hard to find (correct) information

Harvest documents

from all sub-systems

Jul –

Dec

20

10

Hig

h

No

Yes

Clean up Web sites

(using current

systems)

Clean up Web sites

(using a CMS)

CMS Information Systems – Proposed Roadmap