die hard 1 1024 0 die hard 1 1024 0 backward
play

Die Hard 1.1024.0: Die Hard 1.1024.0: Backward compatibility of a - PowerPoint PPT Presentation

May 02, 2023 •493 likes •943 views

4.6.2019 Die Hard 1.1024.0: Backward compatibility of a search engine with persistent IDs Die Hard 1.1024.0: Die Hard 1.1024.0: Backward compatibility of a Backward compatibility of a search engine with persistent search engine with

  1. 4.6.2019 Die Hard 1.1024.0: Backward compatibility of a search engine with persistent IDs Die Hard 1.1024.0: Die Hard 1.1024.0: Backward compatibility of a Backward compatibility of a search engine with persistent search engine with persistent IDs IDs deRSE19 - Conference for Research So�ware Engineers in Germany, 2019-06-04 Thomas Krause (Humboldt-Universität zu Berlin) Stephan Druskat (Friedrich Schiller University Jena, German Aerospace Center (DLR)) 127.0.0.1:5500/die-hard-derse19.html?print-pdf#/title-slide 1/43 1

  2. 4.6.2019 Die Hard 1.1024.0: Backward compatibility of a search engine with persistent IDs Background Background 127.0.0.1:5500/die-hard-derse19.html?print-pdf#/title-slide 2/43 2 . 1

  3. 4.6.2019 Die Hard 1.1024.0: Backward compatibility of a search engine with persistent IDs The Hexatomic project The Hexatomic project “A minimal in�astructure for the sustainable provision of extensible multi-layer annotation so�ware for linguistic corpora ” Funded under the call “Research So�ware Sustainability” issued by DFG under grant number GA 1288/11-1 Runs �om October 2018 until September 2021 127.0.0.1:5500/die-hard-derse19.html?print-pdf#/title-slide 3/43 2 . 2

  4. 4.6.2019 Die Hard 1.1024.0: Backward compatibility of a search engine with persistent IDs The Hexatomic project The Hexatomic project “A minimal in�astructure for the sustainable provision of extensible multi-layer annotation so�ware for linguistic corpora ” Funded under the call “Research So�ware Sustainability” issued by DFG under grant number GA 1288/11-1 Runs �om October 2018 until September 2021 Thomas Krause: computer scientist who slipped into linguistics Stephan Druskat: English M.A. turned so�ware developer & computer scientist Both: Research So�ware Engineers 127.0.0.1:5500/die-hard-derse19.html?print-pdf#/title-slide 4/43 2 . 2

  5. 4.6.2019 Die Hard 1.1024.0: Backward compatibility of a search engine with persistent IDs ANNIS and its query language ANNIS and its query language Web browser-based search and visualization architecture for linguistic corpora with diverse types of annotation . Part of the corpus-tools.org collection of tools for linguists. (Druskat et al. 2016) Annotations are structured information added to text represented as a graph with labels Used by expert users (linguists) to find and analyze linguistic phenomena ANNIS allows finding annotations and combinations of annotations with its domain specific query language AQL AQL describes nodes labels and joins them with operators , which constrain the relation of the nodes in the graph 127.0.0.1:5500/die-hard-derse19.html?print-pdf#/title-slide 5/43 2 . 3

  6. 4.6.2019 Die Hard 1.1024.0: Backward compatibility of a search engine with persistent IDs Semantic Versioning Semantic Versioning Popularized by semver.org (Preston-Werner n.d.) Explicit statement about compatibility between versions of API MAJOR . MINOR . PATCH Only bug fixes when PATCH changes, API does not change Additions to API marked as increase of MINOR Removal and non-backward compatible changes need an increase in MAJOR 127.0.0.1:5500/die-hard-derse19.html?print-pdf#/title-slide 6/43 2 . 4

  7. 4.6.2019 Die Hard 1.1024.0: Backward compatibility of a search engine with persistent IDs Semantic Versioning Semantic Versioning Popularized by semver.org (Preston-Werner n.d.) Explicit statement about compatibility between versions of API MAJOR . MINOR . PATCH Only bug fixes when PATCH changes, API does not change Additions to API marked as increase of MINOR Removal and non-backward compatible changes need an increase in MAJOR Some open questions: What is part of the API in a complex piece of so�ware with multiple components? REST API? �ery language? Data exchange format? User Interface? 127.0.0.1:5500/die-hard-derse19.html?print-pdf#/title-slide 7/43 2 . 4

  8. 4.6.2019 Die Hard 1.1024.0: Backward compatibility of a search engine with persistent IDs Semantic Versioning Semantic Versioning Popularized by semver.org (Preston-Werner n.d.) Explicit statement about compatibility between versions of API MAJOR . MINOR . PATCH Only bug fixes when PATCH changes, API does not change Additions to API marked as increase of MINOR Removal and non-backward compatible changes need an increase in MAJOR Some open questions: What is part of the API in a complex piece of so�ware with multiple components? REST API? �ery language? Data exchange format? User Interface? Do we want to backward-compatible forever? Is there a “1.0 release anxiety”? 127.0.0.1:5500/die-hard-derse19.html?print-pdf#/title-slide 8/43 2 . 4

  9. 4.6.2019 Die Hard 1.1024.0: Backward compatibility of a search engine with persistent IDs Persistent identifiers (PIDs) Persistent identifiers (PIDs) What do I mean exactly when I refer to the “ANNIS so�ware”? http://corpus-tools.org/annis https://github.com/thomaskrause/ANNIS/ ? ? https://github.com/korpling/ANNIS/ ? Version 3.5.1? Version 4? 127.0.0.1:5500/die-hard-derse19.html?print-pdf#/title-slide 9/43 2 . 5

  10. 4.6.2019 Die Hard 1.1024.0: Backward compatibility of a search engine with persistent IDs Persistent identifiers (PIDs) Persistent identifiers (PIDs) What do I mean exactly when I refer to the “ANNIS so�ware”? http://corpus-tools.org/annis https://github.com/thomaskrause/ANNIS/ ? ? https://github.com/korpling/ANNIS/ ? Version 3.5.1? Version 4? I can reference a specific so�ware by a Digital object identifier (DOI): DOI 10.5281/zenodo.1212548 DOI 10.5281/zenodo.1212548 127.0.0.1:5500/die-hard-derse19.html?print-pdf#/title-slide 10/43 2 . 5

  11. 4.6.2019 Die Hard 1.1024.0: Backward compatibility of a search engine with persistent IDs Persistent identifiers (PIDs) Persistent identifiers (PIDs) What do I mean exactly when I refer to the “ANNIS so�ware”? http://corpus-tools.org/annis https://github.com/thomaskrause/ANNIS/ ? ? https://github.com/korpling/ANNIS/ ? Version 3.5.1? Version 4? I can reference a specific so�ware by a Digital object identifier (DOI): DOI 10.5281/zenodo.1212548 DOI 10.5281/zenodo.1212548 In general: resolving an identifier to a resource (digital or not) Should never change , i.e., you can print it in a book! Several systems exist, e.g. DOI, handle.net, … 127.0.0.1:5500/die-hard-derse19.html?print-pdf#/title-slide 11/43 2 . 5

  12. 4.6.2019 Die Hard 1.1024.0: Backward compatibility of a search engine with persistent IDs Persistent identifiers (PIDs) Persistent identifiers (PIDs) What do I mean exactly when I refer to the “ANNIS so�ware”? http://corpus-tools.org/annis https://github.com/thomaskrause/ANNIS/ ? ? https://github.com/korpling/ANNIS/ ? Version 3.5.1? Version 4? I can reference a specific so�ware by a Digital object identifier (DOI): DOI 10.5281/zenodo.1212548 DOI 10.5281/zenodo.1212548 In general: resolving an identifier to a resource (digital or not) Should never change , i.e., you can print it in a book! Several systems exist, e.g. DOI, handle.net, … Some open questions: If a digital resource moves, who updates the reference? Who provides and funds the in�astructure? 127.0.0.1:5500/die-hard-derse19.html?print-pdf#/title-slide 12/43 2 . 5

  13. 4.6.2019 Die Hard 1.1024.0: Backward compatibility of a search engine with persistent IDs Achieving backward compatibility Achieving backward compatibility in ANNIS 4 in ANNIS 4 127.0.0.1:5500/die-hard-derse19.html?print-pdf#/title-slide 13/43 3 . 1

  14. 4.6.2019 Die Hard 1.1024.0: Backward compatibility of a search engine with persistent IDs ANNIS reference links ANNIS reference links ANNIS allows generating short links to query results and single matches, e.g., https://korpling.org/annis3/?id=813c3146-2d10-4d0c-8a1f-1b5efc3c051a Glorified URL shortener: expands to a longer URL encoding the match and the actual query paramters, e.g., https://korpling.org/annis3/#_q=bm9ybT0vZ8O2bm50Lw&_c=UklER0VTX[…] �ery is executed each time the link is opened, no result identifiers are saved 127.0.0.1:5500/die-hard-derse19.html?print-pdf#/title-slide 14/43 3 . 2

  15. 4.6.2019 Die Hard 1.1024.0: Backward compatibility of a search engine with persistent IDs Backward compatibility Backward compatibility Problem: ANNIS 3: AQL queries are mapped to SQL queries and executed by PostgreSQL ANNIS 4: custom in-memory graph-based search engine written in Rust, which directly executes AQL (Krause 2019) All old reference links should still work because the query results are part of the research results. 127.0.0.1:5500/die-hard-derse19.html?print-pdf#/title-slide 15/43 3 . 3

  16. 4.6.2019 Die Hard 1.1024.0: Backward compatibility of a search engine with persistent IDs Backward compatibility Backward compatibility Problem: ANNIS 3: AQL queries are mapped to SQL queries and executed by PostgreSQL ANNIS 4: custom in-memory graph-based search engine written in Rust, which directly executes AQL (Krause 2019) All old reference links should still work because the query results are part of the research results. Users literally printed these links in books. 127.0.0.1:5500/die-hard-derse19.html?print-pdf#/title-slide 16/43 3 . 3

Recommend

The factorization of RSA-1024 D. J. Bernstein University of Illinois at Chicago Abstract: This

The factorization of RSA-1024 D. J. Bernstein University of Illinois at Chicago Abstract: This

The factorization of RSA-1024 D. J. Bernstein University of Illinois at Chicago Abstract: This talk discusses the most important tools for attackers breaking 1024-bit RSA keys today and tomorrow. The same tools will also be useful for

799 views • 47 slides
63 negations Dave Ripley Universities of Connecticut and Melbourne Australasian Association for

63 negations Dave Ripley Universities of Connecticut and Melbourne Australasian Association for

1/ 52 63 negations Dave Ripley Universities of Connecticut and Melbourne Australasian Association for Logic 2013 davewripley@gmail.com 63 negations 2/ 52 1024 1024 to 100 100 to 63 63 davewripley@gmail.com 63 negations 1024 DLL, SM

737 views • 57 slides
Non-Wafer-Scale Sieving Hardware for the NFS: Another Attempt to Cope with 1024-bit Willi

Non-Wafer-Scale Sieving Hardware for the NFS: Another Attempt to Cope with 1024-bit Willi

Non-Wafer-Scale Sieving Hardware for the NFS: Another Attempt to Cope with 1024-bit Willi Geiselmann Rainer Steinwandt The Number Field Sieve Precomputation Relation collection Linear Algebra (Matrix step) Postprocessing

438 views • 15 slides
106 Goudfrooij Figure 1. Schematic architecture of the STIS CCD. The 1024 1024 pixel device

106 Goudfrooij Figure 1. Schematic architecture of the STIS CCD. The 1024 1024 pixel device

2002 HST Calibration Workshop Space Telescope Science Institute, 2002 S. Arribas, A. Koekemoer, and B. Whitmore, eds. Correcting STIS CCD Photometry for CTE Loss 1 Paul Goudfrooij Space Telescope Science Institute, Baltimore, MD 21218, USA

491 views • 10 slides
Microelectrode Arrays for Clinical Mapping: Considerations and Brain Recordings with 1024 Channel

Microelectrode Arrays for Clinical Mapping: Considerations and Brain Recordings with 1024 Channel

Microelectrode Arrays for Clinical Mapping: Considerations and Brain Recordings with 1024 Channel Arrays A subset of slides presented in the symposium has been removed pending publication Shadi A. Dayeh Integrated Electronics and Biointerfaces

398 views • 26 slides
8.07 acres of property located at 1024 W. Workman Ave., West Covina, California 91790, known

8.07 acres of property located at 1024 W. Workman Ave., West Covina, California 91790, known

8.07 acres of property located at 1024 W. Workman Ave., West Covina, California 91790, known generally as the Districts Vincent Childrens Center site } RFP April 2016 no qualified bids } District obtained Waiver from SBE } District

121 views • 8 slides
2018 Investment Conference November 15, 2018 1 1024 South Avenue W, Westfield, NJ 07090 908

2018 Investment Conference November 15, 2018 1 1024 South Avenue W, Westfield, NJ 07090 908

2018 Investment Conference November 15, 2018 1 1024 South Avenue W, Westfield, NJ 07090 908 276 7226 www.rbcpa.com 2 Ronald R. Redfield CPA,PFS started with the firm in 1983 (35 years ago) and became a partner in 1989.

993 views • 68 slides
H H PLAY HARD H HAVE FUN H H NationalAcademyofAthletics.com H H PLAY HARD H HAVE FUN H H THE

H H PLAY HARD H HAVE FUN H H NationalAcademyofAthletics.com H H PLAY HARD H HAVE FUN H H THE

H H PLAY HARD H HAVE FUN H H NationalAcademyofAthletics.com H H PLAY HARD H HAVE FUN H H THE PLAYBOOK H Company Overview H 3 Year plan - Growth Strategy H Problem H Risk Mitigation Milestones H Solution H Competitive Advantage H Market

758 views • 27 slides
HydroCare HC-44 HydroCare HC-44 Hard Water Problems Hard Water Problems Hard Water Costs You

HydroCare HC-44 HydroCare HC-44 Hard Water Problems Hard Water Problems Hard Water Costs You

HydroCare HC-44 HydroCare HC-44 Hard Water Problems Hard Water Problems Hard Water Costs You Money! Hard Water Costs You Money! Limescale originates in the components of the waters carbonic rigidity-calcium carbonate (lime), magnesium

614 views • 8 slides
Programming it's hard to do the programming to get something done details are hard to get

Programming it's hard to do the programming to get something done details are hard to get

Programming it's hard to do the programming to get something done details are hard to get right, very complicated, finicky not enough skilled people to do what is needed therefore, enlist machines to do some of the work leads

130 views • 8 slides
Cache Performance 1 C and cache misses (1) int array[1024]; // 4KB array int even_sum = 0,

Cache Performance 1 C and cache misses (1) int array[1024]; // 4KB array int even_sum = 0,

Cache Performance 1 C and cache misses (1) int array[1024]; // 4KB array int even_sum = 0, odd_sum = 0; for ( int i = 0; i < 1024; i += 2) { even_sum += array[i + 0]; odd_sum += array[i + 1]; } Assume everything but array is kept in

921 views • 67 slides
Hard Problems Some problems are hard to solve. No polynomial time algorithm is known.

Hard Problems Some problems are hard to solve. No polynomial time algorithm is known.

Hard Problems Some problems are hard to solve. No polynomial time algorithm is known. E.g., NP-hard problems such as machine scheduling, bin packing, 0/1 knapsack. Is this necessarily bad? Data encryption relies on difficult

721 views • 15 slides
Recap: Pipes main() { char *s, buf[1024]; int fds[2]; s = Hello World \n"; /* create

Recap: Pipes main() { char *s, buf[1024]; int fds[2]; s = Hello World \n"; /* create

Recap: Pipes main() { char *s, buf[1024]; int fds[2]; s = Hello World \n"; /* create a pipe */ (*) Img. source: http://beej.us/guide/bgipc/output/html/multipage/pipes.html pipe (fds); /* create a new process using fork */ if ( fork

697 views • 32 slides
802.11AX: NEXT GENERATION WIFI 1024 QAM & OFDMA Daan Weller & Raoul Dijksman Arjan van

802.11AX: NEXT GENERATION WIFI 1024 QAM & OFDMA Daan Weller & Raoul Dijksman Arjan van

802.11AX: NEXT GENERATION WIFI 1024 QAM & OFDMA Daan Weller & Raoul Dijksman Arjan van der Vegt (avdvegt@libertyglobal.com), Jan-Willem van Bloem (jvanbloem@libertyglobal.com) WIFI 6 - 802.11AX The 802.11ax amendment focuses on High

635 views • 31 slides
Spectral Extraction of Extended Sources Using Wavelet Interpolation Paul Barrett, Linda Dressel,

Spectral Extraction of Extended Sources Using Wavelet Interpolation Paul Barrett, Linda Dressel,

Spectral Extraction of Extended Sources Using Wavelet Interpolation Paul Barrett, Linda Dressel, STIS Calibration Group 2005 October 26 STIS CCD Characteristics 1024 x 1024 pixels ~ 0.05 arcsecond square pixels PSF = 1/[1 + (x/a)

364 views • 9 slides
10/9/19 Change is Hard! Change Management Principles to Make Hard Changes Simpler. Key

10/9/19 Change is Hard! Change Management Principles to Make Hard Changes Simpler. Key

10/9/19 Change is Hard! Change Management Principles to Make Hard Changes Simpler. Key Take-Aways Change is really, really hard. But, there are ways to simplify the process. Effective change management takes time, commitment, trust

400 views • 10 slides
Improving Hard X-ray Nanoprobe Qingyi Wang, 2010 Lee Teng Internship The Hard X-Ray Nanoprobe

Improving Hard X-ray Nanoprobe Qingyi Wang, 2010 Lee Teng Internship The Hard X-Ray Nanoprobe

Improving Hard X-ray Nanoprobe Qingyi Wang, 2010 Lee Teng Internship The Hard X-Ray Nanoprobe Hard X-Ray Nanoprobe, the laser decoder, and the piezo controller. Its extremely sensitive to operation conditions (noise, vibration, hysteresis,

67 views • 5 slides
[537] Search Engines Tyler Harter 12/10/14 Flash Review Flash Hierarchy Plane : 1024 to 4096

[537] Search Engines Tyler Harter 12/10/14 Flash Review Flash Hierarchy Plane : 1024 to 4096

[537] Search Engines Tyler Harter 12/10/14 Flash Review Flash Hierarchy Plane : 1024 to 4096 blocks - planes accessed in parallel Block : 64 to 256 pages - unit of erase Page : 2 to 8 KB - unit of read and program Block 1111 1111

1.45k views • 106 slides
Why UI testing is SO hard? (and what we can do about it) ivan.inozemtsev@xored.com An abstract

Why UI testing is SO hard? (and what we can do about it) ivan.inozemtsev@xored.com An abstract

Why UI testing is SO hard? (and what we can do about it) ivan.inozemtsev@xored.com An abstract testing tool Control an application-under-test (start/ stop) Manage test resources (create/modify/ exchange) Maintain backward

657 views • 29 slides
Hard Drive to Hard Copy Get your images off the computer and into PRINT! with & Tuesday,

Hard Drive to Hard Copy Get your images off the computer and into PRINT! with & Tuesday,

Hard Drive to Hard Copy Get your images off the computer and into PRINT! with & Tuesday, January 12, 16 I have worked in labs for a collective 20 years and run my own studio since 1996, shooting over 500 weddings and hundreds of corporate

707 views • 23 slides
DIGITAL ANALYTICS IS HARD GET ON WITH IT! Hard learned lessons, If you make any use of tips

DIGITAL ANALYTICS IS HARD GET ON WITH IT! Hard learned lessons, If you make any use of tips

DIGITAL ANALYTICS IS HARD GET ON WITH IT! Hard learned lessons, If you make any use of tips and thoughts this content please give about our digital state. proper credit. I would also love to know! Stphane Hamel #SPWK @SHamel67

758 views • 38 slides
Switch-Less Frequency-Domain Multiplexing of GFET Sensors and Low-Power CMOS Frontend for

Switch-Less Frequency-Domain Multiplexing of GFET Sensors and Low-Power CMOS Frontend for

Switch-Less FDM GFET Sensors LP CMOS Frontend 1024-Ch ECOG Intro FDM ROIC CMOS Chip Conclusions Switch-Less Frequency-Domain Multiplexing of GFET Sensors and Low-Power CMOS Frontend for 1024-Channel ECOG J.Cisneros-Fernndez 1 , M.Dei

517 views • 25 slides
Looking for a simple method to combine hard + 0 jet and hard + 1 jet event generators

Looking for a simple method to combine hard + 0 jet and hard + 1 jet event generators

Looking for a simple method to combine hard + 0 jet and hard + 1 jet event generators Shigeru ODAKA Institute of Particle and Nuclear Studies High Energy Accelerator Research Organization (KEK) shigeru.odaka@kek.jp Motivation

346 views • 19 slides
Regulation in Practice Why is it so hard? Why is it so hard? And what can we do to improve?

Regulation in Practice Why is it so hard? Why is it so hard? And what can we do to improve?

Regulation in Practice Why is it so hard? Why is it so hard? And what can we do to improve? Hanzo van Beusekom Hanzo van Beusekom hanzo@clearconduct.nl Wellington, March 2nd 2014 2 St. Peters mountain in NL (108 m.) St P t t i

383 views • 21 slides

More recommend