Документ взят из кэша поисковой машины. Адрес оригинального документа : http://www.stsci.edu/~carolc/reports/TechSurveyReport_CACv1.pdf Дата изменения: Sat Nov 15 06:29:25 2008 Дата индексирования: Wed Jan 14 10:39:44 2009 Кодировка: Поисковые слова: annular solar eclipse

U.S. DEPARTMENT OF STATE
WASHINGTON, D.C. 20520 18 March 2005 MEMORANDUM UNCLASSIFIED TO: FROM: SUBJECT: CC: MSO IRM/BPC/eDIP - C. Christian ONE Technical Survey Results eDiplomacy

Summary
The OpenNet Everywhere (ONE) solution providing remote access to Department of State (DoS) resources for teleworkers is being made available in 2005. ONE provides access to email, calendar and contacts through Outlook, to the Microsoft Suite, to files and folders through Windows Explorer, and to some other applications as required by users. Core services are through a Citrix server implementation. IRM conducted a technical test of ONE from February 28th through March 11th. Usability testing is scheduled for March 28th through April 8th, depending upon availability of the hardware FOBs required for access to ONE. Approximately 100 technical testers were involved, 60 responded to the survey. In general, ONE technical testers had favorable responses to ONE use. ONE performed satisfactorily for testers with current configurations of browsers and Java on their personal machines. Not surprisingly, problems were encountered with older browsers and Java installations. Most users accessed email (100%), Calendar (83%), the Microsoft Suite (85%), and OpenNet (83%). Over half of the testers accessed files and other applications. Most testers used ONE with broadband connections from home, either a direct connection or through a home LAN connected to broadband. Many users experienced periodic latency in the system and the latency qualification of "acceptable" or "unacceptable" was highly individual. Other problem areas included unexpected dropouts, freezing of the ONE windows, or abrupt closing of all browser windows. Testers commented that users will need significant assistance setting up profiles and mapping shared drives. Testers regularly emailed documents and files back and forth between their DoS email and their personal (e.g., Yahoo!) email accounts so that they could manipulate files on their PC using software they installed and configured.

Detailed Survey Results and Feedback
Testers Technical testers were largely drawn from IRM/OPS and IRM/BPC, including eDiplomacy. Other testers were from DS. Most test respondents were in an IT field (directly supporting 52% and not directly supporting users 32%).


-2Configurations A variety of computer configurations were tried. The most common was Windows XP with Internet Explorer version 6 and Sun Java version 1.4 or 1.5. Other successful configurations included Firefox/Mozilla and Netscape. A few users tried Safari on Mac Powerbooks. Other successful configurations were tested on Windows ME, Windows 2000, Windows NT, Virtual Windows 2000 (on a MAC), an iMAC and MAC OSX. Not surprisingly, access attempts to ONE with configurations documented to be inoperative were, indeed, unsuccessful (access through OpenNet or configurations without Java). In general problems were alleviated if the user downloaded a newer version of Java. Regular users must be urged to download new browser and Java configurations for success. Secure IE did not work with ONE. Occasionally Firefox would close with the following error: "Firefox has generated errors and will be closed by Windows". This did not occur with other Firefox applications. Applications Many testers OpenNet and implemented of the testers connected to and Functions Accessed exercised email, calendar, contacts and the Microsoft Suite. Some testers browsed were able to update websites (Portal X) through ONE. OSIS access had not been in the early stages of the ONE testing, but once enabled worked satisfactorily. 52% did try using other applications from their desktops (outside of ONE), while being ONE, and most attempts were satisfactory.

Testers reported that the message: " Do you want to install and run IE?" comes up in response to access to the ONE IE application. Testers found this confusing and that it had no effect. Users did experience latency with Outlook and, in general experienced that the first application to be invoked took considerable more time than other applications (due to the network connections being enabled?). Testers reported problems with mapping drives and user profiles and recommended that users be assisted through this process as they become ONE users. Most tests with dial-up modems were unsatisfactory in that latency prohibited useful work or caused timeouts. Access Methods Most testers accessed ONE through broad-band connections. Some patient testers were able to use dial-in access. A few users used ONE from Federal Telework centers with success: Laurel, Bowie, Fairfax and Manassas. Access from hotel business centers worked except for those locations that did not have Java installed on public machines. Responsiveness Users found the FOB login to be quick and straightforward. One user had a problem with being locked out of FOB use after 3 failed tries. The lockout period was very long (3 hours). The ONE login was slower and significant latency was experienced especially for the first application launched. Users found that Firefox was more responsive than IE. Some users found that by using windows in full screen mode, performance improved. Responsiveness was judged to be adequate, but for a significant number of users, when latency increased, the system became too frustrating to use.


-3Help Support Most testers did not need help support. Those that did were able to track down individuals they knew who could help (e.g., the SDDI group). Few testers used manuals to troubleshoot problems. Introductory Training Testers easily understood how to use the FOB and login. The introductory material was beneficial, but from the users comments, it appeared that the most important part of the orientation was the hands-on setup period. Some testers still encountered problems with mapped drives and profiles when they began testing at home, so planning for active support while teleworkers get oriented and start using ONE would be beneficial. Many testers did not look at the user guide. Other Issues Testers were pleased to have access mostly to email. Testers also wanted the ability to print and copy/paste to the desktop. · Users were curious why the metaframe would logout in the middle of sessions and seemed to take a while to log back in, but work was rarely lost due to this situation. This will be a point of confusion for teleworkers. · Some users found that when latency occurred or windows were locked, the best procedure was to close ONE down and start over. This latter situation is not a problem if it occurs rarely. · Users should be urged to test their java version using www.javatester.org. · Testers found that the links from the metaframe window did not work. · Testers found that the error message alerting the user that a connection to the Citrix server failed was not helpful. In several instances, ONE was down or had been taken down. It would be helpful to have something on the main ONE page and/or www.state.gov when the ONE servers are down. · Testers noticed that when an application is invoked, sometimes an hourglass icon appears. This icon sometimes "sticks" and does not go away. · Users also had some difficulty using cut/paste functions. · Testers reported that eForms were not viewable, although access to personal accounts and Inbox was functional. · Testers had intermittent success with eRoom. · Testers had difficulty using ePhone.